- Email: [email protected]
Expression of fat mobilizing genes in human epicardial adipose tissue
Atherosclerosis 220 (2012) 122–127
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Expression of fat mobilizing genes in human epicardial adipose tissue I. Jaffer a , M. Riederer b , P. Shah a , P. Peters a , F. Quehenberger c , A. Wood a , H. Scharnagl e , W. März e,f,g , K.M. Kostner d , G.M. Kostner b,∗ a
Princess Alexandra Hospital, Department of Cardiothoratic Surgery, Brisbane, Australia Institute of Molecular Biology and Biochemistry, Medical University of Graz, Harrachgasse 21, Austria Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria d Department of Cardiology, Mater Hospital, Brisbane, Australia e Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Austria f Mannheim Institut für Public Health, Social and Preventive Medicine, University of Heidelberg, Germany g Synlab Academy, Gottlieb Daimler Strasse 25, 68165 Mannheim, Germany b c
a r t i c l e
i n f o
Article history: Received 30 April 2011 Received in revised form 7 October 2011 Accepted 20 October 2011 Available online 2 November 2011 Keywords: Visceral fat RT-PCR Atherosclerosis Lipase Inflammation Bypass patients
a b s t r a c t Background: Epicardial adipose tissue (EAT) mass correlates with Metabolic Syndrome and coronary artery disease (CAD). However, little is known about the expression of genes involved in triglyceride (TG) storage and mobilization in EAT. We therefore analyzed the expression of genes involved in fat mobilization in EAT in comparison to subcutaneous abdominal adipose tissue (AAT) in CAD patients and in controls. Methods: EAT and AAT were obtained during coronary artery bypass graft (CABG) surgery from 16 CAD patients and from 14 non-CAD patients presenting for valve surgery. The state of atherosclerosis was assessed by angiography. RNA from tissues were extracted, reversibly transcribed and quantified by real time polymerase chain reaction (RT-PCR). The following genes were analyzed: perilipin-1 and 5 (PLIN1, PLIN5), lipoprotein lipase (LPL), hormone sensitive lipase (HSL), adipose triglyceride lipase (ATGL), comparative gene identification-58 (CIG-58), angiopoietin like protein 4 (ANGPTL4), in addition to interleukine-6 (IL-6), leptin (LEP) and adiponectin (ADPN). Results: A significant expression of all listed genes could be observed in EAT. The relative expression pattern of the 10 genes in EAT was comparable to the expression in AAT, yet there was a significantly higher overall expression in AAT. The expression of the listed genes was not different between CAD patients and controls. Conclusion: It is suggested that the postulated difference in EAT volume between CAD patients and nonCAD patients is not caused by a differential mRNA expression of fat mobilizing genes. Further work on protein levels and enzyme activities will be necessary to get a complete picture. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction A growing body of evidence suggests that regional fat distribution plays an important part in the development of an unfavorable metabolic and cardiovascular risk profile [1–3]. Increased accumulation of visceral adipose tissue (VAT) is now widely seen as a
Abbreviations: AAT, abdominal adipose tissue; ADPN, adiponectin; ANGPTL4, angiopoietin-like protein 4; ATGL, adipose triglyceride lipase; CAD, coronary artery disease; CABG, coronary artery bypass graft; CIG-58, comparative gene identification-58; CRP, C-reactive protein; EAT, epicardial adipose tissue; IL-6, interleukine-6; LEP, leptin; LPL, lipoprotein lipase; PLIN1, perilipin-1; PLIN5, perilipin-5; qRT-PCR, quantitative real time PCR; SAA, serum amyloid antigen; TG, trigycerides; VAT, visceral adipose tissue. ∗ Corresponding author. Tel.: +43 317 380 7575; fax: +43 316 380 9615. E-mail address: [email protected] (G.M. Kostner). 0021-9150/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2011.10.026
defining characteristic of the Metabolic Syndrome. Despite their similar qualitative properties, different types of adipose tissue, particularly subcutaneous and visceral adipose depots, are now recognized as having distinct quantitative characteristics. While much of the interest has focused on the importance of intra-abdominal VAT, the extra-abdominal visceral fat depots, including epicardial adipose tissue (EAT), have been studied to a lesser extent. VAT in contrast to subcutaneous fat is believed to be metabolically highly active in that hydrolysis and de novo synthesis of triglycerides are continuous processes [reviewed in 4]. There is a lot of information available relating fat mass of different regions with atherosclerosis and myocardial infarction [5–7], yet little is known about the expression of genes involved in TG homeostasis in human EAT. Most of the data on fat mobilization were generated in vitro from cultured adipocytes or from animal models [for a review
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
see Ref. 8]. HSL for decades was believed to be the major if not only lipase responsible for TG hydrolysis from fat depots mediated by the action of stress hormones [9]. Hormonal activation of HSL is triggered by PKA-mediated phosphorylation of several Ser residues. HSL in addition requires several helper proteins for its activation, among them perilipin-1 (PLIN1) and CGI-58 [reviewed in 10]. PLIN1 belongs to the patatin family of proteins and is essential for TG hydrolysis. PLIN1 is also phosphorylated and appears to be mainly responsible for the breakdown of large lipid droplets to form smaller ones. The function of CGI-58 at a molecular level is still not fully explored. Its role in fat mobilization from adipose tissue was delineated from the observation that individuals with mutations in CGI-58 suffer from neutral lipid storage disease, also called Chanarin-Dorfman syndrome [11]. In 2004, Zimmermann et al. [12,13] reported on a new lipase, called adipose triglyceride lipase, ATGL that revolutionized the old dogma of HSL being the predominant neutral TG lipase in adipose tissue. As this research group demonstrated, ATGL in fact is the key TG-hydrolase in adipose tissue as well as in other organs releasing diglycerides, that in turn are further hydrolyzed by HSL. ATGL was shown to interact with CGI-58 as well as perilipin5 (PLIN-5) but with opposite effects: whereas CGI-58 activates TG-hydrolysis PLIN-5 inhibits hydrolysis [14]. Last but not least, adipocytes express another prominent TG-lipase, called lipoprotein lipase (LPL). There is a reciprocal hormonal activation of HSL and LPL, the latter being activated among others by insulin causing an inflow of fatty acids into adipose tissue. LPL activity is inhibited by two closely related proteins, angiopoietin-like protein (ANGPTL)-3 and -4. ANGPTL4 is highly expressed in adipose tissue and has been associated with a variety of diseases. Like LPL, ANGPTL4 is cleaved and inactivated by proprotein convertases such as furin [15]. In the present study we where interested in a potential differential expression of these important enzymes and proteins that regulate fat deposition and mobilization. We therefore analyzed their gene expression in EAT in comparison to AAT by qRT-PCR in 16 patients with obstructive CAD undergoing CABG surgery and in 14 control patients who presented for valve replacement therapy. In addition, the expression of inflammatory markers like interleukin6 (IL-6) and the adipokines adiponectin (ADPN) and leptin (LEP) were measured and correlated with the corresponding plasma levels.
2. Patients and methods 2.1. Subjects studied This project has institutional ethics approval. The patients were recruited at the department of cardiothoracic surgery, Princess Alexandra Hospital, Brisbane, Australia from April to November 2009. Data was prospectively collected and analyzed. All subjects gave their written informed consent before taking part in the study. The study population consisted of two groups. The “CAD” group: included 16 consecutive patients with angiographically proven obstructive CAD who underwent elective primary coronary artery bypass surgery (2 CABG + valve, 1 CABG + cardiac tumor resection). 10 patients had 3-vessel disease, 4 had 2-vessel diseases and the two left main disease. The majority of CAD patients were on statins, aspirin, nitrates, beta blockers and ACE inhibitors. Patients with affected liver disease, renal disease, heart failure class III–IV and patients on fish oil supplements were excluded from the study. The control group had 14 consecutive patients undergoing non-CABG cardiac surgical procedures like aortic and mitral valve replacement (9/14 AVR, 5/15 MVR). These patients had normal coronary angiograms (Table 1). The same exclusion criteria as for the study group applied.
123
Table 1 Pre-operative characterization of patients. All means (SD)
CAD patients
Valve patients
Males/n Age in years BMI Hypertension/n Dyslipidemia/n Diabetes mellitus/n Smoking/n (ex or current)/n
14/16 65 (8.7) 27.81 (3.35) 8/16 10/16 5/16 8/16 (ex) 4/16 (curr) 10/16
9/14 66.47 (12.08) 29.03 (5.80) 8/14 7/14 3/14 5/14 (ex) 1/14(curr)
0.043 0.387 0.237 0.858 0.393 0.491 0.0393
3/14
0.0157
10/16 5/16 3/16 100.34 (9.66) 101.66 (7.98) 0.99 (0.05)
6/14 0/14 5/14 100.10 (14.68) 106.83 (13.00) 0.94 (0.07)
Family history of CAD/n Angina Stable/n Unstable/n Heart failure/n Waist circumfence in cm Hip circumfence in cm W/H ratio
p-Value
<0.001 0.371 0.478 0.094 0.013
There were three distinct components to the research plan. The first stage was the pre-operative stage wherein consecutive patients were screened to assess whether they met the study inclusion criteria and to obtain their informed consent. During this stage, data on age, sex, preoperative risk factors for CAD was prospectively collected (Table 1). Weight (to the nearest 0.1 kg) and height (to the nearest 0.5 cm) was measured while the subjects were fasting and wearing only undergarments. Minimum waist circumference (W, in cm) and maximum hip circumference were measured while the subject was standing with heels together. During the intra-operative phase of the study, patients were prepared for cardiac surgery as per usual. Prior to heparinisation, 20 mL of whole blood was taken and plasma extracted by centrifugation at 3000 rpm for 10 min. All operations were done through a median sternotomy. Prior to bypass, 0.5–1 g fat samples were extracted, EAT was harvested from the anterior surface of the heart along the atrio-ventricular groove and adjacent to the right coronary artery and abdominal fat was obtained from the abdominal subcutaneous fat inferior to the xiphoid at the base of the sternotomy incision. The samples were rinsed of blood, dried, and weighed using a scale sensitive to 1/100 of a gram. Samples were cut into 0.5 g pieces and stored in a container pre-filled with 1 mL Trizol solution (Invitrogen Corp, Carlsbard, CA, USA). Samples were immediately stored in a storage container with dry ice for transport from the operating room. As soon as all samples were collected, the plasma was obtained by centrifugation as described earlier and then all samples were transferred to a −75 ◦ C freezer until final analysis. In the third stage, samples were withdrawn from the −75 ◦ C freezer and packaged immediately on dry ice and shipped via courier to the Department of Biochemistry, University of Graz, Austria for analysis.
3. Sample analyses 3.1. Quantitative real-time PCR (qRT-PCR) RNA of fat pads was isolated using Trizol reagent according to the manufacturer’s protocol. 1.5 g of RNA were reverse transcribed using the Archive cDNA Kit (Applied Biosystems, Foster City, CA) and 0.7 U of a RNAse Inhibitor (Qiagen, Hilden, Germany). The expression of the following genes was analyzed by qRT-PCR: lipoprotein lipase (LPL), hormone-sensitive lipase (HSL), adipose triglyceride lipase (ATGL), comparative gene identification-58 (CGI58) perilipin (PLIN1). RT-PCR analysis was performed in 384-well plates in a total volume of 4 l containing 2 ng of original total RNA.
124
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Analyses were performed as detailed previously [16] using the QuantiFast SYBR green RT-PCR kit (Qiagen) and validated (QuantiTect Primer Assays (Qiagen): LPL: QT00036771; HSL: QT00016093; ATGL: QT00019754; CGI-58: QT00005621; PLIN1: QT00017486; PLIN5: QT01681974; IL-6: QT00083720; LEP: QT00030261; ADPN: QT00055419; ANGPTL4: QT00003661; beta2-microglobulin: QT01665006) according to the manufacturer’s instructions for Light Cycler 480 instruments (Roche Diagnostics). In brief, after the initial heat activation step at 95 ◦ C for 5 min, cycling conditions consisted of 40 cycles of denaturation at 95 ◦ C for 10 s and combined annealing and extension at 60 ◦ C for 30 s. The PCR-efficiency of the target and housekeeping genes was determined by cDNA dilution series prepared from a pooled sample and results were accordingly efficiency corrected with the LightCycler Relative Quantification software (Roche Diagnostics, Basel, Switzerland). mRNA levels of target genes were normalized to human beta-2-microglobulin and expressed as relative ratio (target/reference (ECt )) or as normalized ratio (ECt obtained by division by the respective control group). All samples were assayed in duplicates. 3.1.1. Clinical chemical analyses CRP and IL-6 were measured on a modular automated analyzer (Roche Diagnostics) using reagents and standards from Roche. Other serum proteins were assayed by ELISA: adiponectin (Immundiagnostic, Bensheim, Germany), angiopoietin-like 4 (R&D, Minneapolis, MN, USA), leptin (DRG Instruments GmbH, Marburg, Germany), and SAA (Invitrogen, Camarillo, CA, USA), following the protocols of the manufacturers. 4. Statistical methods Data were analyzed using R 2.12.2 (www.r-project.org). pValues below 0.05 were considered to be statistically significant. Analysis on target/reference values was performed at the logarithmic scale. Outliers in replicates were detected by plotting the quantiles of all pair wise differences against the quantiles of a normal distribution. For outlier removal in technical and biological replicates those measurements with the highest standard deviations were identified and the measurement with the highest absolute residual in a linear model containing patient, target and tissue as factors was removed. In order to test for differences between CABG surgery group and valve replacement group a global test was applied (package global test 5.2.0) [17]. In short, a test statistic was calculated by adding up standardized mean differences (z-values) while taking into account correlations. In the randomization test, p-values were obtained from 10,000 replications. Further insight was obtained from individual z-values and cluster analysis. A nonparametric ANOVA was performed using the package nparLD version 1.3 (www.r-project.org) [16]. The data were rank-transformed and for each of the factors and their interactions an ANOVA-type test statistics was computed. After establishing significance by nonparametric ANOVA, tissues were compared with Wilcoxon’s signed rank test for each target separately [18]. These p-values were multiplied by 10 in order to compensate for multiple testing (Bonferroni correction).
Table 2 Plasma lipid and lipoprotein values of CAD patients and valve patients. All values are means (±SD). Lipid parameter in mmol/L (±SD)
Bypass patients
Valve patients
p-Value
Total cholesterol Triglycerides HDL-cholesterol LDL-cholesterol VLDL-Cholesterol Total/HDL cholesterol
3.77 (1.04) 1.52 (0.84) 0.93 (0.18) 2.16 (0.90) 0.69 (0.37) 4.22 (1.46)
3.90 (1.42) 1.26 (0.85) 1.13 (0.38) 2.57 (1.34) 0.72 (0.61) 3.44 (1.54)
0.834 0.543 0.189 0.488 0.914 0.340
history of CAD and angina pectoris, which were all higher in the CAD group. Table 2 lists the lipid and lipoprotein values of the two groups indicating no significant difference between CAD patients and controls. Even though CAD patients had higher triglyceride and TC/HDL-C and lower HDL-C values, this did not reach statistical significance, likely due to statin therapy in the CAD group. Quantitative RT-PCR of target genes in EAT of all 30 studied individuals revealed significant expression of the lipases and their modulators, LPL, ATGL, HSL, CGI-58 and ANGPTL4 and the lipid droplet proteins PLIN1 and PLIN5, respectively. Notably, expression pattern of these genes in EAT was comparable to that in AAT (Fig. 1). The magnitude of mRNA expression, however, of all genes except for PLIN5 was significantly higher AAT than in EAT calculated by nonparametric ANOVA (p-values below 10−8 ). To elucidate potential patho-physiological implications on gene expression, we compared the expression profile of the target genes involved in TG turnover in EAT of CAD patients with that of controls. As displayed in Fig. 2, both groups exhibited identical expression levels in EAT and in AAT for 6/7 genes. Uncorrected PLIN5 expression was significantly higher in CABG surgery patients; the significance, however, was lost after Bonferroni correction. Since CAD patients reportedly are at a higher oxidative and inflammatory stress we asked the question whether adipokine and inflammatory cytokine levels might be different between CAD patients and controls. As shown in Table 3 there were no differences in plasma concentrations of IL-6, CRP, ADPN, LEP, SAA and ANGPTL4. We also analyzed the gene expression levels of IL-6, ADPN and LEP and ANGPTL4 in EAT and in AAT and correlated them with the corresponding plasma levels of these proteins (Fig. 3). There was a highly significant correlation of ADPN expression in both tissues with plasma ADPN levels. No correlation was found for IL-6, LEP and ANGPL4.
5. Results The CAD patients consisted of 14 males and 2 females and the valve patients (controls) of 8 males and 6 females aged between 55 and 70 years. Mean BMI of all patients was above normal and not significantly different between the 2 groups. The other parameters listed in Table 1 were not significantly different between bypass and controls except for coronary artery stenoses, smoking, family
Fig. 1. Expression of genes involved in triglyceride hydrolysis in 30 individuals (CAD plus valve replacement patients). The results are displayed logarithmically as target/reference ratio. The horizontal line represents the median; outliers exceeding 1.5 fold the length of the box are marked by a circle. Asterisks mark significant p-values after Bonferroni-correction.
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 2. Expression of target genes in abdominal adipose tissue (AAT) and in epicardial adipose tissue (EAT) individually calculated for CAD patients and for valve surgery control patients. The results are displayed logarithmically as target/reference ratio as outlined in Fig. 1.
Table 3 Plasma levels of adipokines and markers for inflammation in CAD and in valve surgery patients. All values are means (±SD). Parameter
Units
CAD
Valve
p-Value Wilcoxon test
IL-6 CRP Adiponectin Leptin SAA Angiopoietin like 4
pg/mL mg/L g/mL ng/mL g/mL ng/mL
4.36 (4.18) 3.77 (4.78) 4.91 (3.04) 13.33 (10.78) 23.29 (16.89) 353.29 (389.74)
3.67 (2.63) 4.93 (10.61) 4.11 (2.63) 6.77 (4.70) 55.93 (64.71) 776.33 (1683.87)
0.950 0.678 0.394 0.129 0.203 0.467
6. Discussion Visceral adipose tissue in general has been recognized as a major source of inflammatory cytokines, which contribute to
125
Metabolic Syndrome and atherosclerotic diseases [19,20]. In addition, it secretes a large amount of adipokines, among them leptin, adiponectin, cylophilin and resistine that have been related to atherogenesis [21,22]. While much of the interest has focused on the importance of intra-abdominal VAT, the extra-abdominal visceral fat depots, including epicardial adipose tissue have been studied less frequently. VAT is metabolically much more active in comparison to subcutaneous fat with hydrolysis going one mainly during night, and de novo triglyceride synthesis post prandially [reviewed in 4]. In EAT, the expression of inflammatory cytokines was reported to be higher than in “tight” adipose tissue and also higher in CAD patients as compared to controls [23]. This is reflected also by the increased expression of genes in EAT involved in oxidative stress [24]. EAT mass has been identified also to correlate with left ventricular function [25] and with congestive heart failure [7,26]. Thus it is not surprising that epicardial fat is an important player in the pathophysiology of CAD [reviewed in 27]. No data on the other hand is available on factors contributing to the generation or mobilization of EAT in humans. ATGL and HSL have been found to be the major enzymes in adipose tissue responsible for neutral lipid hydrolysis [28]. Considering the findings cited above, we hypothesized that these 2 enzymes in addition to other proteins involved in adipose tissue triglyceride homeostasis might reveal a different expression pattern in EAT as compared to subcutaneous AAT and further that differences may exist in the expression profiles of these genes between patients with angiographically proven obstructive CAD and control individuals. EAT and AAT were therefore harvested during surgery from these two patient groups and the mRNA expression of relevant genes was studied by qRT-PCR. As a reference gene we measured the expression of beta-2 microglobulin that exhibited low variance and followed a normal distribution in the fat tissues studied. The highest expression in relation to beta-2 microglobulin in both adipose tissues was found for PLIN1, followed by LPL, HSL and ATGL. PLIN5, CGI-58 and ANGPTL4 were expressed to a significantly lower extent. Of notice, it has to be stated that this inter-target comparison is not absolute as it is based on values generated by relative quantification analyses. The relative expression pattern of the studied genes was not different between EAT and AAT, yet there were significant tissue specific differences in quantity. In fact, all studied genes except for PLIN5 were expressed to a significantly higher degree in AAT than in EAT, indicating a lower metabolic activity in the latter. However, to confirm this hypothesis, correlation of the gene expression profile with protein expression and enzymatic activity (of lipases) will be necessary. Because of the reported higher expression of inflammatory cytokines in EAT in comparison to “tight” adipose tissue [23] we measured IL-6 expression in adipose tissues of our collective. In fact, median values of IL-6 expression in EAT were higher than in AAT (Fig. 4), yet the difference did not reach statistical significance after Bonferroni correction, probably due to the high variation of values with respect to the sample size. In analogy to the expression of genes involved in lipolysis, LEP and ADPN values were also lower in EAT than in AAT (Fig. 4), but again, no differences were seen between CAD patients and valve surgery patients. One interesting finding was that ADPN expression in EAT and in AAT significantly correlated with plasma ADPN values. This was not true for LEP, IL-6 and ANGPTL4. A possible interpretation of these findings might be that ADPN mRNA levels translate to ADPN plasma levels to a higher extent than LEP, ANGPL4 and IL-6 mRNA. For the later, total body fat mass might be more important which has not been measured individually here. Limitations of this study certainly are that we report only on mRNA expression of the different genes and neither protein levels, post translational modifications such as phosphorylation or
126
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 3. (A) Correlation of the expression of cytokines and adipokines in epicardial adipose tissue (EAT) with the corresponding plasma protein levels. CAD-patients and valve surgery patients were combined. (B) Correlation of the expression of cytokines and adipokines in abdominal adipose tissue (AAT) with the corresponding plasma protein levels. CAD-patients and valve surgery patients were combined.
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 4. Expression of the inflammatory marker interlekin-6 (IL-6) and of the adipokines leptin (LEP) and adiponectin (ADPN) in AAT and in EAT of CAD patients and valve surgery control patients. The results are displayed as in Fig. 1.
functional assays and enzymatic activities are reported. Such studies are hardly feasible in humans because of ethical reasons and if so only single proteins might be followed one by one. We consider this manuscript as basis of such further experiments. In summary, we show here for the first time that the most important enzymes catalyzing TG hydrolysis, ATGL, HSL, LPL, as well as the modulators CGI-58 ANGPTL4 are expressed in EAT to a lower extent than in AAT, and that the expression rate is not different between patients with CAD and valve surgery patients. To obtain a complete picture, functional assays of these proteins will be necessary. Disclosure statement The authors have nothing to disclose. Acknowledgements The technical assistance of S. Povoden, M. Tritscher, M. Lechleitner and A. Ibovnik is appreciated. References [1] Mahabadi AA, Massaro JM, Rosito, et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: the Framingham Heart Study. Eur Heart J 2009;30:850–6. [2] Fox CS, Gona P, Hoffmann U, et al. Pericardial fat, intrathoracic fat, and measures of left ventricular structure and function: the Framingham Heart Study. Circulation 2009;119:1586–91.
127
[3] Jain SH, Massaro JM, Hoffmann U, et al. Cross-sectional associations between abdominal and thoracic adipose tissue compartments and adiponectin and resistin in the Framingham Heart Study. Diab Care 2009;32:903–8. [4] Hajer GR, van Haeften TW, Visseren FL. Adipose tissue dysfunction in obesity, diabetes, and vascular diseases. Eur Heart J 2008;29:2959–71. [5] Ohman MK, Wright AP, Wickenheiser KJ, et al. Visceral adipose tissue and atherosclerosis. Curr Vasc Pharmacol 2009;7:169–79. [6] Konishi M, Sugiyama S, Sugamura K, et al. Association of pericardial fat accumulation rather than abdominal obesity with coronary atherosclerotic plaque formation in patients with suspected coronary artery disease. Atherosclerosis 2010;209:573–8. [7] Wanga T-D, Leeb W-J, Fuh-Yuan, et al. Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity. Atherosclerosis 2010;213:279–87. [8] Arner P, Langin D. The role of lipases in human adipose tissue lipolysis. Curr Opin Lipidol 2007;18:246–50. [9] Wong H, Schotz MC. The lipase gene family. J Lipid Res 2002;43:993–9. [10] Zechner R, Kienesberger PC, Haemmerle G, et al. J Lipid Res 2009;50:3–21. [11] Dorfman ML, Hershko C, Eisenberg S, et al. Ichthyosiform dermatosis with systemic lipidosis. Arch Dermatol 1974;110:261–6. [12] Zimmermann R, Strauss JG, Haemmerle G. Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 2004;306(5700):1383–6. [13] Haemmerle G, Lass A, Zimmermann R, et al. Defective lipolysis and altered energy metabolism in mice lacking adipose triglyceride lipase. Science 2006;312(5774):734–7. [14] Wang H, Bell M, Sreenevasan U, et al. Unique regulation of adipose triglyceride lipase (ATGL) by perilipin 5, a lipid droplet-associated protein. J Biol Chem 2011;286:15707–15. [15] Lei X, Shi F, Basu D, et al. Proteolytic processing of angiopoietin-like protein 4 by proprotein convertases modulates its inhibitory effects on lipoprotein lipase activity. J Biol Chem 2011;286:15747–56. [16] Riederer M, Lechleitner M, Hrzenjak A, et al. Endothelial lipase (EL) and ELgenerated lysophosphatidyl cholines promote IL-8 expression in endothelial cells. Atherosclerosis 2011;214:338–44. [17] Goeman JJ, van de Geer SA, de Kort F, et al. A global test for groups of genes: testing association with a clinical outcome. Bioinformatics 2004;20:93–9. [18] Brunner E, Puri ML. Nonparametric methods in factorial designs. Stat Pap 2001;41:1–52. [19] Mazurek T, Zhang L, Zalewski, et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003;08:2460–6. [20] Iacobellis G, Willens HJ. Echocardiographic epicardial fat: a review of research and clinical applications. J Am Soc Echocardiol 2009;22:1311–9. [21] Stofkova A. Leptin and adiponectin: from energy and metabolic dysbalance to inflammation and autoimmunity. Leptin and adiponectin: from energy and metabolic dysbalance to inflammation and autoimmunity. Endocr Regul 2009;43:157–68. [22] Fei J, Cook C, Blough E, Santanama N. Age and sex mediated changes in epicardial fat adipokines. Atherosclerosis 2010;212:488–94. [23] Baker AR, Harte AL, Howell N, et al. Epicardial adipose tissue as a source of nuclear factor-B and c-Jun N-terminal kinase mediated inflammation in patients with coronary artery disease. J Clin Endocrinol Metab 2009;94: 261–7. [24] Salgado-Somoza A, Teijeira-Fernández E, Fernández AL, et al. Proteomic analysis of epicardial and subcutaneous adipose tissue reveals differences in proteins involved in oxidative stress. Am J Physiol Heart Circ Physiol 2010;299: H202–9. [25] Kankaanpaä M, Lehto HR, Pärkkä JP, et al. Myocardial triglyceride content and epicardial fat mass in human obesity: relationship to left ventricular function and serum free fatty acid levels. J Clin Endocrin Metabol 2006;91:4689–95. [26] Doesch C, Haghi D, Flüchter S, et al. Epicardial adipose tissue in patients with heart failure. J Cardiovasc Magn Reson 2010;12:40–9. [27] Sacks HS, Fain JN. Human epicardial adipose tissue: a review. Am Heart J 2007;153:907–17. [28] Schweiger M, Schreiber R, Haemmerle G, et al. Adipose triglyceride lipase and hormone-sensitive lipase are the major enzymes in adipose tissue triacylglycerol catabolism. J Biol Chem 2006;281:40236–41.
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Expression of fat mobilizing genes in human epicardial adipose tissue I. Jaffer a , M. Riederer b , P. Shah a , P. Peters a , F. Quehenberger c , A. Wood a , H. Scharnagl e , W. März e,f,g , K.M. Kostner d , G.M. Kostner b,∗ a
Princess Alexandra Hospital, Department of Cardiothoratic Surgery, Brisbane, Australia Institute of Molecular Biology and Biochemistry, Medical University of Graz, Harrachgasse 21, Austria Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria d Department of Cardiology, Mater Hospital, Brisbane, Australia e Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Austria f Mannheim Institut für Public Health, Social and Preventive Medicine, University of Heidelberg, Germany g Synlab Academy, Gottlieb Daimler Strasse 25, 68165 Mannheim, Germany b c
a r t i c l e
i n f o
Article history: Received 30 April 2011 Received in revised form 7 October 2011 Accepted 20 October 2011 Available online 2 November 2011 Keywords: Visceral fat RT-PCR Atherosclerosis Lipase Inflammation Bypass patients
a b s t r a c t Background: Epicardial adipose tissue (EAT) mass correlates with Metabolic Syndrome and coronary artery disease (CAD). However, little is known about the expression of genes involved in triglyceride (TG) storage and mobilization in EAT. We therefore analyzed the expression of genes involved in fat mobilization in EAT in comparison to subcutaneous abdominal adipose tissue (AAT) in CAD patients and in controls. Methods: EAT and AAT were obtained during coronary artery bypass graft (CABG) surgery from 16 CAD patients and from 14 non-CAD patients presenting for valve surgery. The state of atherosclerosis was assessed by angiography. RNA from tissues were extracted, reversibly transcribed and quantified by real time polymerase chain reaction (RT-PCR). The following genes were analyzed: perilipin-1 and 5 (PLIN1, PLIN5), lipoprotein lipase (LPL), hormone sensitive lipase (HSL), adipose triglyceride lipase (ATGL), comparative gene identification-58 (CIG-58), angiopoietin like protein 4 (ANGPTL4), in addition to interleukine-6 (IL-6), leptin (LEP) and adiponectin (ADPN). Results: A significant expression of all listed genes could be observed in EAT. The relative expression pattern of the 10 genes in EAT was comparable to the expression in AAT, yet there was a significantly higher overall expression in AAT. The expression of the listed genes was not different between CAD patients and controls. Conclusion: It is suggested that the postulated difference in EAT volume between CAD patients and nonCAD patients is not caused by a differential mRNA expression of fat mobilizing genes. Further work on protein levels and enzyme activities will be necessary to get a complete picture. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction A growing body of evidence suggests that regional fat distribution plays an important part in the development of an unfavorable metabolic and cardiovascular risk profile [1–3]. Increased accumulation of visceral adipose tissue (VAT) is now widely seen as a
Abbreviations: AAT, abdominal adipose tissue; ADPN, adiponectin; ANGPTL4, angiopoietin-like protein 4; ATGL, adipose triglyceride lipase; CAD, coronary artery disease; CABG, coronary artery bypass graft; CIG-58, comparative gene identification-58; CRP, C-reactive protein; EAT, epicardial adipose tissue; IL-6, interleukine-6; LEP, leptin; LPL, lipoprotein lipase; PLIN1, perilipin-1; PLIN5, perilipin-5; qRT-PCR, quantitative real time PCR; SAA, serum amyloid antigen; TG, trigycerides; VAT, visceral adipose tissue. ∗ Corresponding author. Tel.: +43 317 380 7575; fax: +43 316 380 9615. E-mail address: [email protected] (G.M. Kostner). 0021-9150/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2011.10.026
defining characteristic of the Metabolic Syndrome. Despite their similar qualitative properties, different types of adipose tissue, particularly subcutaneous and visceral adipose depots, are now recognized as having distinct quantitative characteristics. While much of the interest has focused on the importance of intra-abdominal VAT, the extra-abdominal visceral fat depots, including epicardial adipose tissue (EAT), have been studied to a lesser extent. VAT in contrast to subcutaneous fat is believed to be metabolically highly active in that hydrolysis and de novo synthesis of triglycerides are continuous processes [reviewed in 4]. There is a lot of information available relating fat mass of different regions with atherosclerosis and myocardial infarction [5–7], yet little is known about the expression of genes involved in TG homeostasis in human EAT. Most of the data on fat mobilization were generated in vitro from cultured adipocytes or from animal models [for a review
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
see Ref. 8]. HSL for decades was believed to be the major if not only lipase responsible for TG hydrolysis from fat depots mediated by the action of stress hormones [9]. Hormonal activation of HSL is triggered by PKA-mediated phosphorylation of several Ser residues. HSL in addition requires several helper proteins for its activation, among them perilipin-1 (PLIN1) and CGI-58 [reviewed in 10]. PLIN1 belongs to the patatin family of proteins and is essential for TG hydrolysis. PLIN1 is also phosphorylated and appears to be mainly responsible for the breakdown of large lipid droplets to form smaller ones. The function of CGI-58 at a molecular level is still not fully explored. Its role in fat mobilization from adipose tissue was delineated from the observation that individuals with mutations in CGI-58 suffer from neutral lipid storage disease, also called Chanarin-Dorfman syndrome [11]. In 2004, Zimmermann et al. [12,13] reported on a new lipase, called adipose triglyceride lipase, ATGL that revolutionized the old dogma of HSL being the predominant neutral TG lipase in adipose tissue. As this research group demonstrated, ATGL in fact is the key TG-hydrolase in adipose tissue as well as in other organs releasing diglycerides, that in turn are further hydrolyzed by HSL. ATGL was shown to interact with CGI-58 as well as perilipin5 (PLIN-5) but with opposite effects: whereas CGI-58 activates TG-hydrolysis PLIN-5 inhibits hydrolysis [14]. Last but not least, adipocytes express another prominent TG-lipase, called lipoprotein lipase (LPL). There is a reciprocal hormonal activation of HSL and LPL, the latter being activated among others by insulin causing an inflow of fatty acids into adipose tissue. LPL activity is inhibited by two closely related proteins, angiopoietin-like protein (ANGPTL)-3 and -4. ANGPTL4 is highly expressed in adipose tissue and has been associated with a variety of diseases. Like LPL, ANGPTL4 is cleaved and inactivated by proprotein convertases such as furin [15]. In the present study we where interested in a potential differential expression of these important enzymes and proteins that regulate fat deposition and mobilization. We therefore analyzed their gene expression in EAT in comparison to AAT by qRT-PCR in 16 patients with obstructive CAD undergoing CABG surgery and in 14 control patients who presented for valve replacement therapy. In addition, the expression of inflammatory markers like interleukin6 (IL-6) and the adipokines adiponectin (ADPN) and leptin (LEP) were measured and correlated with the corresponding plasma levels.
2. Patients and methods 2.1. Subjects studied This project has institutional ethics approval. The patients were recruited at the department of cardiothoracic surgery, Princess Alexandra Hospital, Brisbane, Australia from April to November 2009. Data was prospectively collected and analyzed. All subjects gave their written informed consent before taking part in the study. The study population consisted of two groups. The “CAD” group: included 16 consecutive patients with angiographically proven obstructive CAD who underwent elective primary coronary artery bypass surgery (2 CABG + valve, 1 CABG + cardiac tumor resection). 10 patients had 3-vessel disease, 4 had 2-vessel diseases and the two left main disease. The majority of CAD patients were on statins, aspirin, nitrates, beta blockers and ACE inhibitors. Patients with affected liver disease, renal disease, heart failure class III–IV and patients on fish oil supplements were excluded from the study. The control group had 14 consecutive patients undergoing non-CABG cardiac surgical procedures like aortic and mitral valve replacement (9/14 AVR, 5/15 MVR). These patients had normal coronary angiograms (Table 1). The same exclusion criteria as for the study group applied.
123
Table 1 Pre-operative characterization of patients. All means (SD)
CAD patients
Valve patients
Males/n Age in years BMI Hypertension/n Dyslipidemia/n Diabetes mellitus/n Smoking/n (ex or current)/n
14/16 65 (8.7) 27.81 (3.35) 8/16 10/16 5/16 8/16 (ex) 4/16 (curr) 10/16
9/14 66.47 (12.08) 29.03 (5.80) 8/14 7/14 3/14 5/14 (ex) 1/14(curr)
0.043 0.387 0.237 0.858 0.393 0.491 0.0393
3/14
0.0157
10/16 5/16 3/16 100.34 (9.66) 101.66 (7.98) 0.99 (0.05)
6/14 0/14 5/14 100.10 (14.68) 106.83 (13.00) 0.94 (0.07)
Family history of CAD/n Angina Stable/n Unstable/n Heart failure/n Waist circumfence in cm Hip circumfence in cm W/H ratio
p-Value
<0.001 0.371 0.478 0.094 0.013
There were three distinct components to the research plan. The first stage was the pre-operative stage wherein consecutive patients were screened to assess whether they met the study inclusion criteria and to obtain their informed consent. During this stage, data on age, sex, preoperative risk factors for CAD was prospectively collected (Table 1). Weight (to the nearest 0.1 kg) and height (to the nearest 0.5 cm) was measured while the subjects were fasting and wearing only undergarments. Minimum waist circumference (W, in cm) and maximum hip circumference were measured while the subject was standing with heels together. During the intra-operative phase of the study, patients were prepared for cardiac surgery as per usual. Prior to heparinisation, 20 mL of whole blood was taken and plasma extracted by centrifugation at 3000 rpm for 10 min. All operations were done through a median sternotomy. Prior to bypass, 0.5–1 g fat samples were extracted, EAT was harvested from the anterior surface of the heart along the atrio-ventricular groove and adjacent to the right coronary artery and abdominal fat was obtained from the abdominal subcutaneous fat inferior to the xiphoid at the base of the sternotomy incision. The samples were rinsed of blood, dried, and weighed using a scale sensitive to 1/100 of a gram. Samples were cut into 0.5 g pieces and stored in a container pre-filled with 1 mL Trizol solution (Invitrogen Corp, Carlsbard, CA, USA). Samples were immediately stored in a storage container with dry ice for transport from the operating room. As soon as all samples were collected, the plasma was obtained by centrifugation as described earlier and then all samples were transferred to a −75 ◦ C freezer until final analysis. In the third stage, samples were withdrawn from the −75 ◦ C freezer and packaged immediately on dry ice and shipped via courier to the Department of Biochemistry, University of Graz, Austria for analysis.
3. Sample analyses 3.1. Quantitative real-time PCR (qRT-PCR) RNA of fat pads was isolated using Trizol reagent according to the manufacturer’s protocol. 1.5 g of RNA were reverse transcribed using the Archive cDNA Kit (Applied Biosystems, Foster City, CA) and 0.7 U of a RNAse Inhibitor (Qiagen, Hilden, Germany). The expression of the following genes was analyzed by qRT-PCR: lipoprotein lipase (LPL), hormone-sensitive lipase (HSL), adipose triglyceride lipase (ATGL), comparative gene identification-58 (CGI58) perilipin (PLIN1). RT-PCR analysis was performed in 384-well plates in a total volume of 4 l containing 2 ng of original total RNA.
124
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Analyses were performed as detailed previously [16] using the QuantiFast SYBR green RT-PCR kit (Qiagen) and validated (QuantiTect Primer Assays (Qiagen): LPL: QT00036771; HSL: QT00016093; ATGL: QT00019754; CGI-58: QT00005621; PLIN1: QT00017486; PLIN5: QT01681974; IL-6: QT00083720; LEP: QT00030261; ADPN: QT00055419; ANGPTL4: QT00003661; beta2-microglobulin: QT01665006) according to the manufacturer’s instructions for Light Cycler 480 instruments (Roche Diagnostics). In brief, after the initial heat activation step at 95 ◦ C for 5 min, cycling conditions consisted of 40 cycles of denaturation at 95 ◦ C for 10 s and combined annealing and extension at 60 ◦ C for 30 s. The PCR-efficiency of the target and housekeeping genes was determined by cDNA dilution series prepared from a pooled sample and results were accordingly efficiency corrected with the LightCycler Relative Quantification software (Roche Diagnostics, Basel, Switzerland). mRNA levels of target genes were normalized to human beta-2-microglobulin and expressed as relative ratio (target/reference (ECt )) or as normalized ratio (ECt obtained by division by the respective control group). All samples were assayed in duplicates. 3.1.1. Clinical chemical analyses CRP and IL-6 were measured on a modular automated analyzer (Roche Diagnostics) using reagents and standards from Roche. Other serum proteins were assayed by ELISA: adiponectin (Immundiagnostic, Bensheim, Germany), angiopoietin-like 4 (R&D, Minneapolis, MN, USA), leptin (DRG Instruments GmbH, Marburg, Germany), and SAA (Invitrogen, Camarillo, CA, USA), following the protocols of the manufacturers. 4. Statistical methods Data were analyzed using R 2.12.2 (www.r-project.org). pValues below 0.05 were considered to be statistically significant. Analysis on target/reference values was performed at the logarithmic scale. Outliers in replicates were detected by plotting the quantiles of all pair wise differences against the quantiles of a normal distribution. For outlier removal in technical and biological replicates those measurements with the highest standard deviations were identified and the measurement with the highest absolute residual in a linear model containing patient, target and tissue as factors was removed. In order to test for differences between CABG surgery group and valve replacement group a global test was applied (package global test 5.2.0) [17]. In short, a test statistic was calculated by adding up standardized mean differences (z-values) while taking into account correlations. In the randomization test, p-values were obtained from 10,000 replications. Further insight was obtained from individual z-values and cluster analysis. A nonparametric ANOVA was performed using the package nparLD version 1.3 (www.r-project.org) [16]. The data were rank-transformed and for each of the factors and their interactions an ANOVA-type test statistics was computed. After establishing significance by nonparametric ANOVA, tissues were compared with Wilcoxon’s signed rank test for each target separately [18]. These p-values were multiplied by 10 in order to compensate for multiple testing (Bonferroni correction).
Table 2 Plasma lipid and lipoprotein values of CAD patients and valve patients. All values are means (±SD). Lipid parameter in mmol/L (±SD)
Bypass patients
Valve patients
p-Value
Total cholesterol Triglycerides HDL-cholesterol LDL-cholesterol VLDL-Cholesterol Total/HDL cholesterol
3.77 (1.04) 1.52 (0.84) 0.93 (0.18) 2.16 (0.90) 0.69 (0.37) 4.22 (1.46)
3.90 (1.42) 1.26 (0.85) 1.13 (0.38) 2.57 (1.34) 0.72 (0.61) 3.44 (1.54)
0.834 0.543 0.189 0.488 0.914 0.340
history of CAD and angina pectoris, which were all higher in the CAD group. Table 2 lists the lipid and lipoprotein values of the two groups indicating no significant difference between CAD patients and controls. Even though CAD patients had higher triglyceride and TC/HDL-C and lower HDL-C values, this did not reach statistical significance, likely due to statin therapy in the CAD group. Quantitative RT-PCR of target genes in EAT of all 30 studied individuals revealed significant expression of the lipases and their modulators, LPL, ATGL, HSL, CGI-58 and ANGPTL4 and the lipid droplet proteins PLIN1 and PLIN5, respectively. Notably, expression pattern of these genes in EAT was comparable to that in AAT (Fig. 1). The magnitude of mRNA expression, however, of all genes except for PLIN5 was significantly higher AAT than in EAT calculated by nonparametric ANOVA (p-values below 10−8 ). To elucidate potential patho-physiological implications on gene expression, we compared the expression profile of the target genes involved in TG turnover in EAT of CAD patients with that of controls. As displayed in Fig. 2, both groups exhibited identical expression levels in EAT and in AAT for 6/7 genes. Uncorrected PLIN5 expression was significantly higher in CABG surgery patients; the significance, however, was lost after Bonferroni correction. Since CAD patients reportedly are at a higher oxidative and inflammatory stress we asked the question whether adipokine and inflammatory cytokine levels might be different between CAD patients and controls. As shown in Table 3 there were no differences in plasma concentrations of IL-6, CRP, ADPN, LEP, SAA and ANGPTL4. We also analyzed the gene expression levels of IL-6, ADPN and LEP and ANGPTL4 in EAT and in AAT and correlated them with the corresponding plasma levels of these proteins (Fig. 3). There was a highly significant correlation of ADPN expression in both tissues with plasma ADPN levels. No correlation was found for IL-6, LEP and ANGPL4.
5. Results The CAD patients consisted of 14 males and 2 females and the valve patients (controls) of 8 males and 6 females aged between 55 and 70 years. Mean BMI of all patients was above normal and not significantly different between the 2 groups. The other parameters listed in Table 1 were not significantly different between bypass and controls except for coronary artery stenoses, smoking, family
Fig. 1. Expression of genes involved in triglyceride hydrolysis in 30 individuals (CAD plus valve replacement patients). The results are displayed logarithmically as target/reference ratio. The horizontal line represents the median; outliers exceeding 1.5 fold the length of the box are marked by a circle. Asterisks mark significant p-values after Bonferroni-correction.
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 2. Expression of target genes in abdominal adipose tissue (AAT) and in epicardial adipose tissue (EAT) individually calculated for CAD patients and for valve surgery control patients. The results are displayed logarithmically as target/reference ratio as outlined in Fig. 1.
Table 3 Plasma levels of adipokines and markers for inflammation in CAD and in valve surgery patients. All values are means (±SD). Parameter
Units
CAD
Valve
p-Value Wilcoxon test
IL-6 CRP Adiponectin Leptin SAA Angiopoietin like 4
pg/mL mg/L g/mL ng/mL g/mL ng/mL
4.36 (4.18) 3.77 (4.78) 4.91 (3.04) 13.33 (10.78) 23.29 (16.89) 353.29 (389.74)
3.67 (2.63) 4.93 (10.61) 4.11 (2.63) 6.77 (4.70) 55.93 (64.71) 776.33 (1683.87)
0.950 0.678 0.394 0.129 0.203 0.467
6. Discussion Visceral adipose tissue in general has been recognized as a major source of inflammatory cytokines, which contribute to
125
Metabolic Syndrome and atherosclerotic diseases [19,20]. In addition, it secretes a large amount of adipokines, among them leptin, adiponectin, cylophilin and resistine that have been related to atherogenesis [21,22]. While much of the interest has focused on the importance of intra-abdominal VAT, the extra-abdominal visceral fat depots, including epicardial adipose tissue have been studied less frequently. VAT is metabolically much more active in comparison to subcutaneous fat with hydrolysis going one mainly during night, and de novo triglyceride synthesis post prandially [reviewed in 4]. In EAT, the expression of inflammatory cytokines was reported to be higher than in “tight” adipose tissue and also higher in CAD patients as compared to controls [23]. This is reflected also by the increased expression of genes in EAT involved in oxidative stress [24]. EAT mass has been identified also to correlate with left ventricular function [25] and with congestive heart failure [7,26]. Thus it is not surprising that epicardial fat is an important player in the pathophysiology of CAD [reviewed in 27]. No data on the other hand is available on factors contributing to the generation or mobilization of EAT in humans. ATGL and HSL have been found to be the major enzymes in adipose tissue responsible for neutral lipid hydrolysis [28]. Considering the findings cited above, we hypothesized that these 2 enzymes in addition to other proteins involved in adipose tissue triglyceride homeostasis might reveal a different expression pattern in EAT as compared to subcutaneous AAT and further that differences may exist in the expression profiles of these genes between patients with angiographically proven obstructive CAD and control individuals. EAT and AAT were therefore harvested during surgery from these two patient groups and the mRNA expression of relevant genes was studied by qRT-PCR. As a reference gene we measured the expression of beta-2 microglobulin that exhibited low variance and followed a normal distribution in the fat tissues studied. The highest expression in relation to beta-2 microglobulin in both adipose tissues was found for PLIN1, followed by LPL, HSL and ATGL. PLIN5, CGI-58 and ANGPTL4 were expressed to a significantly lower extent. Of notice, it has to be stated that this inter-target comparison is not absolute as it is based on values generated by relative quantification analyses. The relative expression pattern of the studied genes was not different between EAT and AAT, yet there were significant tissue specific differences in quantity. In fact, all studied genes except for PLIN5 were expressed to a significantly higher degree in AAT than in EAT, indicating a lower metabolic activity in the latter. However, to confirm this hypothesis, correlation of the gene expression profile with protein expression and enzymatic activity (of lipases) will be necessary. Because of the reported higher expression of inflammatory cytokines in EAT in comparison to “tight” adipose tissue [23] we measured IL-6 expression in adipose tissues of our collective. In fact, median values of IL-6 expression in EAT were higher than in AAT (Fig. 4), yet the difference did not reach statistical significance after Bonferroni correction, probably due to the high variation of values with respect to the sample size. In analogy to the expression of genes involved in lipolysis, LEP and ADPN values were also lower in EAT than in AAT (Fig. 4), but again, no differences were seen between CAD patients and valve surgery patients. One interesting finding was that ADPN expression in EAT and in AAT significantly correlated with plasma ADPN values. This was not true for LEP, IL-6 and ANGPTL4. A possible interpretation of these findings might be that ADPN mRNA levels translate to ADPN plasma levels to a higher extent than LEP, ANGPL4 and IL-6 mRNA. For the later, total body fat mass might be more important which has not been measured individually here. Limitations of this study certainly are that we report only on mRNA expression of the different genes and neither protein levels, post translational modifications such as phosphorylation or
126
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 3. (A) Correlation of the expression of cytokines and adipokines in epicardial adipose tissue (EAT) with the corresponding plasma protein levels. CAD-patients and valve surgery patients were combined. (B) Correlation of the expression of cytokines and adipokines in abdominal adipose tissue (AAT) with the corresponding plasma protein levels. CAD-patients and valve surgery patients were combined.
I. Jaffer et al. / Atherosclerosis 220 (2012) 122–127
Fig. 4. Expression of the inflammatory marker interlekin-6 (IL-6) and of the adipokines leptin (LEP) and adiponectin (ADPN) in AAT and in EAT of CAD patients and valve surgery control patients. The results are displayed as in Fig. 1.
functional assays and enzymatic activities are reported. Such studies are hardly feasible in humans because of ethical reasons and if so only single proteins might be followed one by one. We consider this manuscript as basis of such further experiments. In summary, we show here for the first time that the most important enzymes catalyzing TG hydrolysis, ATGL, HSL, LPL, as well as the modulators CGI-58 ANGPTL4 are expressed in EAT to a lower extent than in AAT, and that the expression rate is not different between patients with CAD and valve surgery patients. To obtain a complete picture, functional assays of these proteins will be necessary. Disclosure statement The authors have nothing to disclose. Acknowledgements The technical assistance of S. Povoden, M. Tritscher, M. Lechleitner and A. Ibovnik is appreciated. References [1] Mahabadi AA, Massaro JM, Rosito, et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: the Framingham Heart Study. Eur Heart J 2009;30:850–6. [2] Fox CS, Gona P, Hoffmann U, et al. Pericardial fat, intrathoracic fat, and measures of left ventricular structure and function: the Framingham Heart Study. Circulation 2009;119:1586–91.
127
[3] Jain SH, Massaro JM, Hoffmann U, et al. Cross-sectional associations between abdominal and thoracic adipose tissue compartments and adiponectin and resistin in the Framingham Heart Study. Diab Care 2009;32:903–8. [4] Hajer GR, van Haeften TW, Visseren FL. Adipose tissue dysfunction in obesity, diabetes, and vascular diseases. Eur Heart J 2008;29:2959–71. [5] Ohman MK, Wright AP, Wickenheiser KJ, et al. Visceral adipose tissue and atherosclerosis. Curr Vasc Pharmacol 2009;7:169–79. [6] Konishi M, Sugiyama S, Sugamura K, et al. Association of pericardial fat accumulation rather than abdominal obesity with coronary atherosclerotic plaque formation in patients with suspected coronary artery disease. Atherosclerosis 2010;209:573–8. [7] Wanga T-D, Leeb W-J, Fuh-Yuan, et al. Association of epicardial adipose tissue with coronary atherosclerosis is region-specific and independent of conventional risk factors and intra-abdominal adiposity. Atherosclerosis 2010;213:279–87. [8] Arner P, Langin D. The role of lipases in human adipose tissue lipolysis. Curr Opin Lipidol 2007;18:246–50. [9] Wong H, Schotz MC. The lipase gene family. J Lipid Res 2002;43:993–9. [10] Zechner R, Kienesberger PC, Haemmerle G, et al. J Lipid Res 2009;50:3–21. [11] Dorfman ML, Hershko C, Eisenberg S, et al. Ichthyosiform dermatosis with systemic lipidosis. Arch Dermatol 1974;110:261–6. [12] Zimmermann R, Strauss JG, Haemmerle G. Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 2004;306(5700):1383–6. [13] Haemmerle G, Lass A, Zimmermann R, et al. Defective lipolysis and altered energy metabolism in mice lacking adipose triglyceride lipase. Science 2006;312(5774):734–7. [14] Wang H, Bell M, Sreenevasan U, et al. Unique regulation of adipose triglyceride lipase (ATGL) by perilipin 5, a lipid droplet-associated protein. J Biol Chem 2011;286:15707–15. [15] Lei X, Shi F, Basu D, et al. Proteolytic processing of angiopoietin-like protein 4 by proprotein convertases modulates its inhibitory effects on lipoprotein lipase activity. J Biol Chem 2011;286:15747–56. [16] Riederer M, Lechleitner M, Hrzenjak A, et al. Endothelial lipase (EL) and ELgenerated lysophosphatidyl cholines promote IL-8 expression in endothelial cells. Atherosclerosis 2011;214:338–44. [17] Goeman JJ, van de Geer SA, de Kort F, et al. A global test for groups of genes: testing association with a clinical outcome. Bioinformatics 2004;20:93–9. [18] Brunner E, Puri ML. Nonparametric methods in factorial designs. Stat Pap 2001;41:1–52. [19] Mazurek T, Zhang L, Zalewski, et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003;08:2460–6. [20] Iacobellis G, Willens HJ. Echocardiographic epicardial fat: a review of research and clinical applications. J Am Soc Echocardiol 2009;22:1311–9. [21] Stofkova A. Leptin and adiponectin: from energy and metabolic dysbalance to inflammation and autoimmunity. Leptin and adiponectin: from energy and metabolic dysbalance to inflammation and autoimmunity. Endocr Regul 2009;43:157–68. [22] Fei J, Cook C, Blough E, Santanama N. Age and sex mediated changes in epicardial fat adipokines. Atherosclerosis 2010;212:488–94. [23] Baker AR, Harte AL, Howell N, et al. Epicardial adipose tissue as a source of nuclear factor-B and c-Jun N-terminal kinase mediated inflammation in patients with coronary artery disease. J Clin Endocrinol Metab 2009;94: 261–7. [24] Salgado-Somoza A, Teijeira-Fernández E, Fernández AL, et al. Proteomic analysis of epicardial and subcutaneous adipose tissue reveals differences in proteins involved in oxidative stress. Am J Physiol Heart Circ Physiol 2010;299: H202–9. [25] Kankaanpaä M, Lehto HR, Pärkkä JP, et al. Myocardial triglyceride content and epicardial fat mass in human obesity: relationship to left ventricular function and serum free fatty acid levels. J Clin Endocrin Metabol 2006;91:4689–95. [26] Doesch C, Haghi D, Flüchter S, et al. Epicardial adipose tissue in patients with heart failure. J Cardiovasc Magn Reson 2010;12:40–9. [27] Sacks HS, Fain JN. Human epicardial adipose tissue: a review. Am Heart J 2007;153:907–17. [28] Schweiger M, Schreiber R, Haemmerle G, et al. Adipose triglyceride lipase and hormone-sensitive lipase are the major enzymes in adipose tissue triacylglycerol catabolism. J Biol Chem 2006;281:40236–41.