Introduction

Obesity has been considered one of the important risk factors in the development of metabolic syndrome, insulin resistance, diabetes mellitus as well as atherosclerotic cardiovascular diseases.1, 2 Cumulative evidence supports the notion that adipose tissue, in addition to storing excess energy in the form of lipid, may play an active role via its endocrine function in systemic vascular inflammation.3, 4 A variety of pro- and anti-inflammatory mediators and cytokines, which were expressed and secreted from adipose tissue, collectively termed as ‘adipocytokines’, have been demonstrated to participate in the various stages of atherogenesis, ranging from endothelial dysfunction to plaque destabilization and rupture.5

The pathogenic profiles of central visceral or peripheral adipose tissues have been investigated.6, 7 The differences in gene expression and protein concentrations of adipocyte-specific molecules from these two distinct fat distributions raised the concern that intrinsic disparities of specific fat depots may exist.8 Furthermore, regional epicardial fat tissue have recently been demonstrated to influence pathogenically the development of coronary artery diseases (CAD).9, 10, 11 The anatomical and microscopic findings in which epicardial fat tissue have direct contact with underlying myocardium without fascia structures support the potential paracrine effects of epicardial adipocytokines on myocardial metabolism and CAD pathogenesis.12 Subsequent clinical studies have also noted a strong correlation between the epicardial fat mass and associated risk of cardiovascular disease.13, 14

Adiponectin and leptin were exclusively identified in mature adipocytes.15 Tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), although classified in the family of ‘adipocytokines’, have been shown to be secreted primarily from infiltrated macrophages within the adipose tissue.15 The expression and secretion of leptin by adipocytes can be induced by IL-6 and inhibited by TNF-α, and adiponectin can inhibit TNF-α-induced changes in monocyte adhesion molecule expression and the endothelial inflammatory responses.16, 17, 18 These emphasize the interaction among adipocytokines and their release by fat cells. In contrast to the markedly increased levels of leptin, TNF-α and IL-6 in obesity, the level of adiponectin is negatively correlated with body mass index. Moreover, hypoadiponectinemia is associated with the presence of insulin resistance, type 2 diabetes and risk of myocardial infarction.19, 20 Visfatin, a newly recognized adipocytokine secreted by visceral fat, was found to have insulin-mimetic properties and exert hypoglycemic effects.21 The expressions of visfatin in epicardial adipose tissue remain unknown.

Although previous investigations have pointed out that tissue expressions of these adipocytokines might influence the cardiovascular risk profile, no reports to date make direct comparisons of these representative adipocytokines between epicardial and abdominal fat depots in patients with and without CAD. Moreover, local inflammatory burden in these adipose tissues may not correlate well with plasma levels of circulating adipocytokines.9 In the context of autocrine, paracrine and endocrine functions as well as site-specific characteristics of these adipose tissues, we sought to evaluate and compare tissue concentrations of adiponectin, TNF-α, IL-6, leptin and visfatin derived from epicardial and abdominal fat tissues in CAD and non-CAD subjects.

Methods

Study population

Between January 2004 and July 2005, a total of 46 patients who underwent elective coronary artery bypass grafting (CABG) surgery were enrolled in this study. Another 12 patients who received open-heart surgery (including valvular replacement (n=9), ventricular (n=2) and atrial (n=1) septal defects) without significant coronary stenosis in the pre-operative coronary angiographic examination were enrolled as the control groups. We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research. This study was approved by the Institutional Review Board of Kaohsiung Medical University and all patients provided written informed consent.

Blood collection

On the morning of surgery, peripheral venous blood was drawn into pyrogen-free tubes with ethylenediaminetetraacetic acid as an anticoagulant. Serum glucose, lipid profiles and other biochemical profiles were analyzed in the Kaohsiung Medical University Hospital central laboratory.

Adipose tissue collection

All adipose tissue biopsy samples were obtained prior to initiation of cardiopulmonary bypass pumping. Epicardial adipose tissue biopsy samples (average 1.0–1.5 g) were harvested adjacent to the proximal right coronary artery. Abdominal adipose tissue biopsy samples were taken from central abdominal fat depots. After removal of visible blood vessels, all fat tissue biopsy specimens were rinsed with phosphate-buffered saline and cut into small pieces (2 mm3) in a 12-well plate. According to tissue weight, serum-free Dulbecco's modified Eagle's media (DMEM, 2 ml g−1) was added to the well and incubated at 37 °C in a CO2 incubator with gentle rocking. After 3 h, the conditioned media were collected and centrifuged at 4 °C for 10 min. The supernatants from adipose tissue cultures were stored in aliquots at −76 °C for measurement of released inflammatory mediators and adipocytokines by enzyme-linked immunosorbent assay (ELISA), as described previously.9

Enzyme immunoassay for tissue inflammatory mediators and adipocytokines

Adipose tissue inflammatory mediators and adipokines (conditioned medium) were assayed by ELISA kits according to the manufacturer's recommended procedure. Tissue levels of adiponectin and visfatin were determined by ELISA kits (Phoenix Pharmaceuticals, Belmont, CA, USA) and leptin by ELISA kits (Assaypro, Brooklyn, NY, USA). Tissue IL-6 and TNF-α levels were quantified with human immunoassay (R&D Systems, Minneapolis, MN, USA).

Statistical analysis

Demographic and laboratory data are presented as mean±s.d. Adipocytokines in fat tissues are found with broader range and therefore displayed in mean±s.e.m. Comparisons between CAD and control groups were calculated by independent two-sample t-test and comparisons between abdominal and epicardial adipose tissues in the study or control groups were analyzed by paired t-test. The P-value <0.05 was considered statistically significant. Statistical analysis was performed by SPSS 11.0 software (Chicago, IL, USA).

Results

Baseline characteristics

The demographic and clinical characteristics at baseline between patients with CAD (CAD groups) and non-CAD control subjects (non-CAD groups) are summarized in Table 1. The patients in the CAD group were significantly older than those in non-CAD group. Patients with hypertension and previous history of dyslipidemia were significantly prevalent in the CAD group relative to the non-CAD group. However, the lipid profiles including levels of total cholesterol, low- and high-density lipoprotein cholesterol and triglyceride were not significantly different in these two groups when enrolled. No significant difference in the measurement of waist circumference and body mass index was demonstrated between these two groups. Left ventricular ejection fraction was quite similar in the two groups. The majority of patients (93.5%) who underwent CABG surgery in the CAD group had triple-vessel disease. One patient with single-vessel disease received valve replacement surgery due to severe aortic stenosis with symptoms of heart failure; the other two patients with two-vessel disease underwent CABG operation because of significant stenotic lesions in left main trunk. The percentage of antiplatelet, angiotensinogen-converting enzyme inhibitor/angiotensin receptor blocker and statin therapy in CAD group were significantly higher than in non-CAD control subjects (P<0.05).

Table 1 Demographic and clinical characteristics of CAD and non-CAD groups
Full size table

Laboratory profiles of adipocytokines concentration

All the laboratory data were obtained in the morning prior to surgery and are also summarized in Table 2. No significant differences were demonstrated between CAD and non-CAD groups in terms of lipid and sugar profiles. There were also no significant differences in hemoglobin A1C and prevalence of diabetes in these two groups.

Table 2 Biochemical profiles in CAD and non-CAD groups
Full size table

Comparisons of tissue adipocytokine levels in epicardial and abdominal fat tissues between the CAD and non-CAD groups

The tissue concentrations of adiponectin, IL-6, leptin and visfatin in two anatomically distinct fat tissues (epicardial and abdominal, respectively) were compared between CAD and non-CAD groups. The absolute values were presented in Table 3 and illustrated in Figure 1. The tissue levels of adiponectin in abdominal and epicardial fat depots in the CAD group were significantly reduced. Conversely, the tissue levels of TNF-α and IL-6 in both of the abdominal and epicardial fat depots were significantly elevated in the CAD group relative to those in the non-CAD group. Significantly higher tissue levels of leptin and visfatin in abdominal and epicardial fat depots were also observed.

Table 3 Summary of tissue adipocytokines and proinflammatory markers in the CAD and non-CAD groups
Full size table
Figure 1
figure 1

Concentrations of adipocytokines ((a) adiponectin, (b) tumor necrosis factor (TNF)-α, (c) interleukin-6 (IL-6); (d) leptin and (e) visfatin) derived from abdominal and epicardial fat tissues in patients with coronary artery disease (coronary artery diseases (CAD), n=46) and control subjects (non-CAD, n=12). The concentration of IL-6, leptin and visfatin were significantly higher in all selected adiposity from CAD patients than control subjects. In addition, significantly higher levels of these adipocytokines were found in abdominal, followed by epicardial adiposity in CAD patients. Conversely, tissue adiponectin levels were significantly reduced in all selected adiposity from CAD patients and the levels were the lowest in abdominal followed by epicardial adiposity.

Full size image

Comparison of the levels of adipocytokines from epicardial versus abdominal fat tissue in the CAD group

Tissue adipocytokines from abdominal and epicardial fat depots in the CAD group were shown in Table 3 and Figure 1. The tissue adiponectin level was significantly lower in abdominal fat than in the epicardial fat depots (P<0.001). Conversely, the tissue levels of IL-6, TNF-α, leptin and visfatin were significantly higher in abdominal fat depots than in epicardial fat depots.

Discussion

The major findings of this study were as follows: (1) In patients with critical CAD, who underwent CABG surgery, tissue concentrations of adipocytokines (TNF-α, IL-6, leptin and visfatin) from abdominal and epicardial adipose tissues were significantly higher than those in the non-CAD control subjects. In addition, tissue concentrations of adiponectin were significantly reduced in the CAD patients. (2) In patients with critical CAD, abdominal adipose tissue exhibited significantly higher expression of adipocytokines (TNF-α, IL-6, leptin and visfatin) than epicardial adipose tissue. Adiponectin concentration was significantly lower in abdominal fat than that in epicardial fat.

The pathogenic profiles of adipocytokines expressed in fat tissues can alter the vascular inflammation burden and have remote influences on the cardiovascular system.3, 7 The impact of adipocytokines has been recently extended from endocrine to paracrine roles.8 Epicardial adipose tissues lack fascial structures to separate themselves from underlying myocardium and the phenomenon with increased infiltration of inflammatory cells (macrophages, T cells, mast cells) in epicardial adipose tissue had been observed in CAD patients. The epicardial fat thickness was found to be correlated with the severity of CAD in patients with CAD.22 These reports suggested that adipocytokines derived from epicardial fat might have direct proinflammatory responses in CAD pathogenesis.9, 10, 15, 23, 24 However, while epicardial fat is totally absent in congenital generalized lipodystrophy, coronary atherosclerosis can still occur.23 The epicardial fat amount correlates with heart weight but the presence of myocardial ischemia and hypertrophy does not alter the ratio of epicardial fat to cardiac muscle mass.24 Therefore, the debate still existed that whether local impact of epicardial fat to CAD is more important than systemic influence, which largely depended on abdominal fat.

Human epicardial adipose tissue was initially postulated as a source of inflammatory mediators by Mazurek et al.9 and their study showed higher mRNA and protein expression of TNF-α, IL-6, MCP-1 and IL1β from epicardial than subcutaneous adipose tissue in patients with critical CAD. Later, a study by Iacobellis et al. showed that adiponectin protein expression in epicardial adipose tissue by western analysis was significantly lower in patients with critical CAD than non-CAD subjects.10 Recently, Baker et al. investigated mRNA expression of several key adipocytokines in epicardial adipose tissue from CAD patients who underwent CABG in comparison with those in abdominal subcutaneous, omental and thigh adipose tissues from non-CAD, body mass index matched control subjects.11 They also demonstrated the potential contribution of macrophage infiltration by showing higher CD45 mRNA expression in epicardial fat depots from CAD patients. All these studies, however, still have not addressed the potential difference and relative significance among anatomically distinct adipose tissues in terms of either critical CAD patients or non-CAD control subjects because of the lack of parallel study designs.

In patients with critical CAD who underwent CABG surgery in the present study, significantly higher abundance of proinflammatory mediators, including TNF-α and IL-6, and significantly lower concentrations of anti-inflammatory adipocytokines, adiponectin, were consistently observed in epicardial and abdominal adipose tissues compared with non-CAD control subjects. Notably, these highly significant differences were demonstrated even while significantly higher percentage of patients in CAD group taking anti-platelet agents, angiotensinogen-converting enzyme inhibitors or angiotensin receptor blocker as well as statin compared with those in control groups. These results were compatible with previous studies and supported the notion that TNF-α and IL-6 might elicit detrimental effects and adiponectin might elicit beneficial effects in the pathogenesis of CAD.4 In addition, to the best of our knowledge, the present study is the first to demonstrate that tissue concentrations of visfatin in epicardial and abdominal fat depots were both significantly elevated in patients with critical CAD than in non-CAD control subjects. Although upregulated in obesity, visfatin has been postulated to be beneficial because of its insulin-mimetic and hypoglycemic effects.21 Further studies are necessary to elucidate the controversial role of visfatin and its beneficial or detrimental effects on the pathophysiology of CAD.

To determine possible regional differences of fat tissue and adipocytokine expression, fat depots of epicardial and abdominal adipose tissues were harvested in each patient with critical CAD in our study. In a direct comparison between these two anatomically distinct sites of fat tissues, the findings revealed that the abdominal adiposity store exhibited significantly higher expressions of TNF-α, IL-6, leptin and visfatin than epicardial adipose tissue. The adiposity stores of adiponectin was significantly lower in abdominal than in epicardial adipose tissue. In considering the relatively greater fat biomass of abdominal adipose tissue compared with that of the epicardial tissue, the contributions of adipocytokines from abdominal fats on systemic inflammation might be speculated to be more significant.5 However, although abdominal adiposity contributes significantly to adipocytokine levels, the epicardial fat tissue may exert direct effects on coronary atherosclerosis via their paracrine function. In fact, measurements of abdominal adiposity by magnetic resonance imaging and, recently, epicardial fat mass by echocardiography have both been correlated with cardiovascular risks.12, 13, 14 Further studies may still be necessary to delineate the relative contribution and significance of epicardial and abdominal adiposites on the pathogenesis of CAD.

Study limitations

The present study has several limitations. First, most patients who underwent CABG surgery in our study were diagnosed with critical triple-vessel CAD, which explained the significant differences of tissue adipocytokines levels compared with non-CAD patients. Whether these findings can be applied to CAD patients with less severity needs to be interpreted with caution. Second, the mRNA expression of these adipocytokines was not performed in these two anatomically distinct fat depots in the present study. However, the measurements of protein concentrations of these adipocytokines adjusted by tissue weight directly reflect the activities of bioactive substances and are easier to interpret. Finally, the sample size might be modest for association studies to achieve statistical significance.

In conclusion, the tissue concentrations of adipocytokines derived from epicardial or abdominal fat tissue in our study exhibited significantly different profiles in patients with critical CAD versus non-CAD subjects. Higher tissue levels of TNF-α, IL-6, leptin and visfatin and lower tissue levels of adiponectin were demonstrated in patients with critical CAD. Between the two distinct sites of adipose tissues, expression of adipocytokines was more prominent in abdominal than epicardial fat, which may suggest that the abdominal adiposity may play the more significant role in the pathogenesis of systemic atherosclerosis.