Adiponectin levels and expression of adiponectin receptors in isolated monocytes from overweight patients with coronary artery disease

Background Adiponectin has insulin-sensitizing and anti-atherosclerotic effects, partly mediated through its action on monocytes. We aimed to determine adiponectin levels and expression of its receptors (AdipoR1 and AdipoR2) in peripheral monocytes from overweight and obese patients with coronary artery disease (CAD). Methods Fifty-five overweight/obese patients, suspected for CAD, underwent coronary angiography: 31 were classified as CAD patients (stenosis ≥ 50% in at least one main vessel) and 24 as nonCAD. Quantitative RT-PCR and flow cytometry were used for determining mRNA and protein surface expression of adiponectin receptors in peripheral monocytes. A high sensitivity multiplex assay (xMAP technology) was used for the determination of plasma adiponectin and interleukin-10 (IL-10) secreted levels. Results Plasma adiponectin levels were decreased in CAD compared to nonCAD patients (10.9 ± 3.1 vs. 13.8 ± 5.8 μg/ml respectively, p = 0.033). In multivariable analysis, Matsuda index was the sole independent determinant of adiponectin levels. AdipoR1 and AdipoR2 protein levels were decreased in monocytes from CAD compared to nonCAD patients (59.5 ± 24.9 vs. 80 ± 46 and 70.7 ± 39 vs. 95.6 ± 47.8 Mean Fluorescence Intensity Arbitrary Units respectively, p < 0.05). No significant differences were observed concerning the mRNA levels of the adiponectin receptors between CAD and nonCAD patients. AdipoR2 protein levels were positively correlated with plasma adiponectin and Matsuda index (r = 0.36 and 0.31 respectively, p < 0.05 for both). Furthermore, basal as well as adiponectin-induced IL-10 release was reduced in monocyte-derived macrophages from CAD compared to nonCAD subjects. Conclusions Overweight patients with CAD compared to those without CAD, had decreased plasma adiponectin levels, as well as decreased surface expression of adiponectin receptors in peripheral monocytes. This fact together with the reduced adiponectin-induced IL-10 secretion from CAD macrophages could explain to a certain extent, an impaired atheroprotective action of adiponectin.


Background
Cardiovascular disease (CVD) is a leading cause of mortality worldwide accounting for 30% of total global deaths and the number one leading cause of deaths in developing countries. Various studies have demonstrated the association of obesity and insulin resistance with cardiovascular risk. It has been recognized that inflammatory mechanisms play a pivotal role in initiation, maintenance and progress of CVD and that the association of CVD with obesity is mediated through the adipose tissue secretory activities.
Adiponectin, an adipose tissue secreted protein, has been well recognized to exhibit insulin-sensitizing, antiinflammatory and anti-atherosclerotic properties, which are mediated through its receptors, AdipoR1 and AdipoR2 [1,2]. These receptors are ubiquitously expressed in most organs as well as in human peripheral monocytes, and in monocyte-derived macrophages [2][3][4][5]. In these cells, adiponectin has been shown to modulate their inflammatory activity and most importantly to inhibit their transformation to foam cells, a hallmark of atherosclerosis [6][7][8][9][10]. More specifically, adiponectin induces the anti-inflammatory cytokine IL-10 in human monocytes and macrophages [5,6,11], while it suppresses the LPS-stimulated release of IL-6 in porcine macrophages [8]. Furthermore, adiponectin-induced secretion of IL-6 and IL-8 is reduced in monocytes from patients with type 1 or type 2 diabetes [5,12]. This fact may have significant implications in terms of atherosclerotic processes, since these cells play a pivotal role in inflammation and atherosclerosis [13].
Adiponectin levels are known to be decreased in patients with obesity, type 2 diabetes and coronary artery disease (CAD) [14][15][16][17]. However, in recent studies, adiponectin was shown to be associated with an adverse outcome in patients with CAD, challenging the emerging evidence for its role [18][19][20][21]. In addition, alterations at the level of adiponectin receptors expression have been reported but with inconclusive results showing upregulation or downregulation in the presence of insulin resistance [5,[22][23][24][25][26][27][28]. Recently, a lower abundance of adiponectin receptors was found in monocytes from overweight/obese patients with type 2 diabetes [5]. However, the expression levels of adiponectin receptors have not been examined in CAD.
The aim of this study was to examine whether the relative mRNA and protein (surface) expression of Adi-poR1 and AdipoR2 in human peripheral monocytes is altered in overweight/obese patients with CAD, and whether this might relate to their circulating adiponectin levels and to indices of insulin resistance and atherosclerosis.

Subjects and investigation
Our study included patients with Body Mass Index (BMI) above 25 Kg/m 2 who underwent elective coronary angiography for the investigation of the existence of chronic stable CAD. Patients with a 50% or greater diameter stenosis in at least one major coronary artery were considered as CAD positive patients. The exclusion criteria were: unstable angina or acute myocardial infarction, unstable condition including infections, heart failure, malignancies, renal disease (creatinin level > 1.5 mg/dl). All patients underwent intima-media thickness (IMT) assessment in 2 paired segments of both common carotids, as well as in carotid bulbs using Bmode ultrasound imaging. IMT was measured using an automatically trading software (Vivid 7 GE Horton, Norway). The research conducted in this paper has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association and the study protocol was approved by the Scientific and Ethics Committee of "Attikon" University General Hospital. All patients gave their written informed consent.
Two to three weeks after their discharge from the Cardiology Department, patients were subjected to a 2-hour oral glucose (75gr) tolerance test (OGTT) for the assessment of their insulin resistance status. All subjects were informed to halt their medication at least one day before the OGTT, as necessary. Insulin resistance was calculated by the homeostasis model assessment index (HOMA-IR) as [fasting insulin (μU/mL) × fasting glucose (mmol/L)]/ 22.5 [29]. Insulin sensitivity was calculated by the Matsuda index as [10.000/square root of (fasting glucose × fasting insulin) × (mean glucose during OGTT × mean insulin during OGTT)] [30]. At the beginning of the OGTT, fasting blood was drawn and plasma was kept at -80°C for measurement of adiponectin levels. Fasting triglycerides, total and HDL-cholesterol were determined by an ILAB analyser (Instrumentation Laboratory SpA, Italy). Plasma glucose levels at fasting and during the OGTT, were analyzed on a Falcor 300 Chemical Analyser (Menarini Diagnostics, Italy), while plasma insulin levels were measured by IRMA (INSI-CTK, DiaSorin, Italy), in all individuals.
BMI was calculated as the ratio of body weight (Kg) per height (m 2 ). Waist and hip circumference, weight and height were also measured in all individuals. Fiftyfive patients, aged between 41 and 75 years, were eligible to be included in the study among 71 subjects screened. All patients were of a stable weight and had been on a normal isocaloric diet with normal physical activity, not participating in any specific exercise program. Information about use of medication, such as antihyperlipidemic, antidiabetic and antihypertensive medication was also collected. None of the patients were taking thiazolidinedione medication.

Study of adiponectin receptors surface (protein) expression with flow cytometry
The mononuclear cell fraction was isolated from peripheral blood and the surface adiponectin receptors, Adi-poR1 and AdipoR2, were determined after staining the cells with suitable antibodies. More specifically, the antibodies used were rabbit anti-mouse AdipoR1 and rabbit anti-mouse AdipoR2 (Alpha Diagnostic International, San Antonio, USA), each corresponding to the extracellular carboxy-terminus domain of the respective receptor. Since the antibodies were not fluorochrome conjugated, they were labelled with the Zenon™ Alexa Fluor ® 488 Rabbit IgG labelling kit (Invitrogen, Carlsbad, CA, USA). Cells were incubated for 30 min with Alexa Fluor ® 488-conjugated immunoglobulin in a ratio 1 × 10 6 cells μg -1 of immunoglobulin under mild constant shaking. The monocyte fraction was simultaneously stained with anti-CD14-PE monoclonal antibody (BD Biosciences, San Jose, CA, USA). After incubation with the antisera, cells were fixed with 0.1% (w/v) paraformaldehyde. A histogram of log green fluorescence of each adiponectin receptor was used for the determination of the mean fluorescence intensity of each sample ( Figure 1).
The specificity of the antibodies used was evaluated by staining cells with isotype controls suitable for each antisera and the blockage of Fc-receptors prior to staining [31].
Two-colour flow cytometric analysis was performed on a BD FACSCalibur 4 colour flow cytometer (BD Biosciences) as previously described [31]. Data acquisition and analysis were performed using the BD CELL Quest Pro software (BD Biosciences). Results are expressed in mean fluorescence intensity arbitrary units (MFI AU).

Isolation of peripheral monocytes
Monocytes were isolated from mononuclear cells using a Magnetic Cell-Sorting technique with CD14 Human MicroBeads (Miltenyi Biotec, Gladbach, Germany). The exact number of cells was determined cytometrically with Flow-Count™ Flurospheres (Beckman-Coulter, Miami, USA). The purity of the magnetically isolated CD14 + monocytes was determined after staining of the cells with a CD14 PE-CD45 FITC antibody (BD Biosciences, USA)

RNA preparation
Total RNA was extracted from CD14 + monocytes using the Tripure Isolation Reagent (Roche Diagnostics GmbH, Mannheim, Germany), and the integrity of RNA samples was determined on ethidium bromide-agarose gels and spectrophotometrically. For the removal of any residual DNA contamination, RNA samples were preincubated with RQ1 RNase-free DNase (Promega, Madison, WI, USA) before the reverse transcription reaction. Reverse transcription of 1 μg of total RNA was performed in all samples with the Transcriptor First Strand cDNA Synthesis Kit using random hexamer primers, according to the manufacturer's instructions (Roche Diagnostics GmbH, Mannheim, Germany).

Relative real-time PCR quantification of adiponectin receptors mRNA expression
A quantitative real-time PCR using fluorescently labeled hybridization probes was developed for detecting relative AdipoR1 and AdipoR2 mRNA levels in human CD14 + monocytes, in a spectrofluorometric thermal cycler (LightCycler, ROCHE, Manheim, Germany) as previously described [32].
Relative quantification is the method that determines the changes in steady-state mRNA levels of a gene  At the end of the incubation period, cell supernatants were collected, aliquoted and kept at -80°C for measurement of released IL-10.

Immunoassays
Circulating levels of adiponectin were measured in the plasma of all subjects, using a high sensitivity multiplex assay (xMAP technology) and the fluorescently labeled microsphere beads (HCVD1-67AK Lincoplex kit, Millipore Corp., MA, USA), in a LUMINEX 200 instrument (Luminex Corp., USA). Sensitivity of the assay was 56.0 pg/ml, intra-assay and inter-assay coefficients of variation were 9.2% and 15.9%, respectively. All samples were diluted 1:100 and measured in duplicate. Samples that could not be detected, were measured again using a higher dilution.
IL-10 secreted levels were measured in the supernatants from human monocyte-derived macrophages using a MILLIPLEX MAP kit (MPXHCYTO-60 K, Millipore Corp., MA, USA), in a LUMINEX 200 instrument (Luminex Corp., USA). Sensitivity of the assay was 0.5 pg/ml and intra-assay and inter-assay coefficients of variation were 5.2% and 9.5%.

Statistical analysis
Statistical analysis was performed using the SPSS version 14.0.1 software (Chicago, Ill, USA). Data are expressed as mean ± SD. Variables that were not normally distributed (AdipoR1 and AdipoR2 mRNA levels, adiponectin, HOMA index) were logarithmically transformed. Chi square was implemented for categorical variables. Parametric statistical analyses (Pearson's correlation test, unpaired or paired t-test) were applied where appropriate. Multivariate regression analysis (results as mean ± SE), as well as analysis of covariance (ANCOVA) were also performed if any adjustments were necessary. A pvalue of less than 0.05 was considered statistically significant.

Patients' characteristics
The clinical and metabolic characteristics, as well as the therapeutic regimen of the patients included in the study are presented in Table 1. From the 55 subjects, 46 were males and no differences were observed in BMI between males and females (28.9 ± 2.6 vs. 28.2 ± 2.1 respectively, p = 0.44). Overweight/obese patients with CAD were more likely to be males, while no differences were observed with regard to age, BMI, lipid profile, systolic and diastolic blood pressure compared to similar weight patients without CAD. Patients with CAD had higher values of waist to hip ratio, higher carotid IMT values, decreased levels of plasma adiponectin and decreased insulin sensitivity as assessed by Matsuda index (Table 1). Moreover, the decrease in plasma adiponectin levels remained significant after adjustment for gender (p = 0.046) and BMI (p = 0.025).

Plasma adiponectin levels and associations
Circulating plasma adiponectin levels were positively associated with Matsuda index and HDL-cholesterol levels, overall (r = 0.315, p = 0.022 and r = 0.29 p = 0.048, respectively, Table 2). In stepwise multivariable analysis with age, gender, waist to hip ratio, the presence of CAD, Matsuda index and the various medications as independent variables, insulin sensitivity (Matsuda index) (beta ± SE: 0.43 ± 0.02, p = 0.024) was the sole independent determinant of adiponectin levels (R 2 = 0.10).

Expression of adiponectin receptors
Both AdipoR1 and AdipoR2 were found to be expressed in peripheral monocytes at mRNA and protein surface level. While AdipoR1 mRNA was more abundant compared with AdipoR2 mRNA in monocytes (3.5 ± 3.2 vs. 1 ± 0.9 AU, p < 0.001), the opposite was true for the protein surface expression levels (68.2 ± 36.5 vs. 81.3 ± 44.3 MFI AU respectively, p = 0.002). Overweight/obese patients with CAD, apart from the decreased plasma adiponectin levels, had decreased AdipoR1 and AdipoR2 protein surface levels compared to patients without CAD, while they did not show any difference with respect to the levels of AdipoR1 and AdipoR2 mRNA (Figure 2A and 2B). The difference in protein surface expression of AdipoR1 and AdipoR2 with respect to the presence of CAD remained significant for AdipoR1 after adjustment for gender difference (p = 0.033) but was attenuated for AdipoR2 (p = 0.074). However, after adjustment for Matsuda index the difference in protein surface expression was attenuated for both AdipoR1 and AdipoR2 (p = 0.13 and 0.08 respectively). Moreover, a robust positive correlation was observed between AdipoR1 and AdipoR2 mRNA, as well as between AdipoR1 and AdipoR2 surface levels ( Figure 3). AdipoR1 mRNA levels were inversely correlated with Matsuda index, whereas AdipoR2 mRNA levels were correlated with systolic blood pressure and triglycerides ( Table 2). AdipoR2 protein surface levels were associated with plasma adiponectin levels and Matsuda index ( Figure 4) and retained this association with plasma adiponectin levels (beta ± SE: 107.58 ± 40.7, p = 0.011) and Matsuda index (beta ± SE: 13.21 ± 6.46, p = 0.047) in multiple regression analysis independent of age, gender and waist to hip ratio. When the latter analysis was confined to the subgroup of CAD patients, AdipoR2 protein surface levels were also correlated with plasma adiponectin levels (beta ± SE: 110.02 ± 48.89, p = 0.034) and Matsuda index (beta ± SE: 25.603 ± 6.060, p < 0.001) independent of age, gender and waist to hip ratio.
Although no associations were revealed overall, between common carotid or bulb IMT with adiponectin receptors mRNA or protein surface levels ( Table 2), in subgroup analyses, a positive correlation was found between AdipoR2 protein levels and carotid bulb IMT in CAD patients (r = 0.428, p = 0.026).

Plasma adiponectin levels and associations
Adiponectin is an adipocyte derived protein that exerts anti-inflammatory and anti-atherogenic effects in part by modulating monocyte and macrophage functions. More specifically, adiponectin may prime human monocytes toward the anti-inflammatory M2 phenotype, inhibit the secretion of inflammatory cytokines from monocyte-derived macrophages, prevent monocyte adhesion to endothelial cells and suppress macrophageto-foam cell transformation [6][7][8][9][10]. In addition, the new adiponectin paralogous protein CTRP-3 has been reported to play a significant role by inhibiting LPSinduced IL-6 and TNF release from human monocytes [33]. Adiponectin receptors have been detected in human monocytes and macrophages [3,4] and recently it was found that variants of the AdipoR2 could be a determinant for atherosclerosis independent of insulin resistance [34]. However the significance of the adiponectin receptors in mediating its beneficial effects in states of CAD has not been examined.
In this study, adiponectin levels were found to be marginally decreased in overweight patients with CAD compared to similar weight control individuals. Although this finding could be expected according to the evidence from previous studies [14], this view is questioned. A couple of studies failed to reveal a relationship of adiponectin levels with early [35] or advanced atherosclerosis [36]. In other studies, adiponectin levels were found to be associated with an adverse outcome in patients with CAD [18][19][20][21]. In addition, Rizza et al. reported that adiponectin levels in patients with CAD were not correlated with the severity of CAD, but inversely associated with defects of glucose metabolism [37]. Interestingly, in our study, when adjusting for several factors, adiponectin levels did not differ with respect to the presence of CAD, but were independently associated with Matsuda index. Furthermore, the positive association of plasma adiponectin with HDL-C levels is in agreement with previous epidemiological reports [38,39] and more recent ones suggesting that the apparent correlation of adiponectin with HDL-C is at least partly mediated by a mechanism whereby adiponectin increases the cholesterol efflux from macrophages and thus preventing atherosclerosis [40].

Expression of adiponectin receptors and adiponectin effects
We confirmed the presence of AdipoR1 and AdipoR2 at both mRNA and protein surface level in peripheral monocytes from subjects with BMI above normal. The robust correlation between mRNA or surface expression levels of the two receptors found, could indicate a synchronous regulation of both genes at transcriptional and translational level. It should be reported that a high correlation between these two receptors at mRNA level has been previously reported in peripheral monocytes from subjects with type 2 diabetes and normal controls, as well as in human skeletal muscle cells [5,26].
However, mRNA and surface expression levels did not exhibit a similar pattern of expression judging from their absolute values and their correlations. In fact, AdipoR1 mRNA levels were higher compared with AdipoR2 respective ones, while the opposite was true for the surface expression levels. In addition, Adi-poR1 mRNA levels were inversely correlated with Matsuda index and AdipoR2 mRNA were positively correlated with systolic blood pressure and triglycerides, while surface expression levels of AdipoR2 were positively correlated with Matsuda index and adiponectin levels. Moreover, the presence of CAD was associated with decreased surface expression of both receptors. Although adiponectin receptors are expressed in cells within atherosclerotic carotid plaques [3], we did not uncover any association between adiponectin receptors and carotid IMT values. Yet, in CAD patients, AdipoR2 surface expression levels were correlated with carotid bulb IMT. This finding could represent a compensatory mechanism to decreased adiponectin plasma levels. The aforementioned results point out that the translation and the transcription process of the adiponectin receptors may not follow the same direction in states of insulin resistance and atherosclerosis. The decreased surface expression levels in combination with the decreased adiponectin levels in CAD could contribute to a vicious circle of decreased adiponectin action and aggravated atherosclerotic processes.
One of the anti-inflammatory actions of the adiponectin is the induction of IL-10 [6]. In our in vitro experiments, IL-10 basal protein production was lower in monocyte-derived macrophages from CAD patients compared to that of nonCAD. This fact could be partly attributed to the constant exposure to lower adiponectin plasma levels and to basal reduced AdipoR1 and Adi-poR2 surface levels in these cells from CAD patients. Moreover, adiponectin-induced IL-10 secretion was higher in macrophages from nonCAD compared to CAD subjects, although the effect of adiponectin was significant in both CAD and nonCAD cells. Expression of human adiponectin receptors has been examined mainly at mRNA level only by a limited number of studies which have yielded inconclusive results. Chinetti et al reported that the expression level of Adi-poR1 in monocytes decreases during their differentiation into macrophages but remains higher than that of Adi-poR2 [3]. However, only AdipoR2 was upregulated by PPAR (peroxisome proliferators-activated receptors) agonists which have insulin-sensitizing properties [3]. These results are in line with our findings, since we also reported a higher level of AdipoR1 mRNA in human monocytes and we found a positive correlation of Adi-poR2 surface expression levels with insulin sensitivity. Moreover, protein levels of both adiponectin receptors in monocytes from overweight/obese diabetic patients were reported to be decreased compared to normal weight control subjects [5]. Expression of mRNA levels of adiponectin receptors in human adipose tissue, in states of insulin resistance, has been reported to be either decreased [23][24][25] or increased [27]. A similar inconsistency has been reported for muscle, where mRNA levels are positively associated with insulin sensitivity [28] or with insulin resistance [22].

Study limitations
Our study results should be interpreted with regards to some limitations. Firstly, we did not include a control group of normal weight subjects. Nevertheless, this does not invalidate our findings, since the aim of the present study was to examine adiponectin receptors with regard only to the presence of CAD and to correlate their expression with indices of insulin resistance and atherosclerosis. The later became feasible, as our subjects did not differ in terms of BMI values but exhibited a satisfactory variability in insulin sensitivity status, allowing various correlations to be performed. However, it should be noted that CAD patients presented higher values of waist to hip ratio compared to nonCAD patients, suggesting an increased prevalence of visceral adiposity which in turn is a main determinant of adiponectin levels and insulin sensitivity status. In addition, central obesity is associated with increased levels of saturated nonesterified fatty acids, which in combination with hyperinsulinemia, may activate human monocytes to produce proinflammatory cytokines and support the development and propagation of the subacute, chronic inflammatory state [41]. Interestingly, when adjusting for Matsuda index, the difference in the surface expression levels of the adiponectin receptors between CAD and nonCAD patients was attenuated. Moreover, Adi-poR2 surface levels in the subgroup of CAD patients were correlated with adiponectin, suggesting that the decrease in AdipoR2 levels in monocytes from CAD patients may be in relation to the decrease in insulin sensitivity. Another limitation of this study is that our patients were not subjected to a satisfactory wash-out period concerning their therapeutic regimen. However, treatment was interrupted at least 24 hours before the measurements and adjustments for drug categories were taken into account in our analyses. Lastly, a limitation of our study could be the relatively small population studied; this could imply a reduced statistical power.

Conclusions
In summary our data provide evidence that mRNA and surface expression of the adiponectin receptors in human monocytes seems to be subjected to different regulatory mechanisms in states of insulin resistance and atherosclerosis. Furthermore, the latter states are associated with lower adiponectin plasma levels and a decrease in surface expression of adiponectin receptors in monocytes. This, in turn could aggravate the antiatherogenic action of adiponectin in CAD patients. Research and Technology (GSRT) (25%) and the private sector within the framework of the Operational Programme, "Competitiveness" (Measure 8.3) of the 3 rd Community Support Programme. We would like to thank Ms V. Frangaki, R.N., for nursing help, and Ms A. Koukourava for excellent technical assistance.