Introduction

There is strong evidence that type 2 diabetes is associated with increased oxidative stress [1], which may contribute to the elevated risk of atherothrombosis. Oxidatively modified LDL are implicated in the pathogenesis of atherosclerosis and could be present in plasma from type 2 diabetic patients. On the one hand, autoantibodies to oxidised LDL were found to be increased in type 2 diabetic patients with macrovascular disease [2]. On the other hand, we recently showed that LDL from type 2 diabetic patients contained increased levels of lipid peroxides [3]. There is controversy on the occurrence of oxidative stress in patients with the metabolic syndrome. Some studies have shown higher levels of circulating oxidised LDL [4–6] and isoprostanes [7] in individuals with the metabolic syndrome. Other studies show no difference in oxidised LDL levels [8] or markers of oxidative damage (plasma F2-isoprostanes, hydroxy-eicosatetraenoic acid [HETE]) [9] between patients with the metabolic syndrome and healthy individuals. In addition, most studies have focused on lipid peroxides issued from the non-enzymatic peroxidation of arachidonic acid [7, 9]. Paradoxically, no data (to our knowledge) are available on circulating lipid peroxides issued from linoleic acid, although it is the main polyunsaturated fatty acid (PUFA) in LDL. Considering that the metabolic syndrome increases the risk of developing type 2 diabetes and cardiovascular diseases [10, 11], and that oxidative stress plays an important role in type 2 diabetes, it is important to determine whether oxidised LDL are already present in plasma from patients with the metabolic syndrome.

Increased platelet activation occurs in type 2 diabetes [12] and the metabolic syndrome [13, 14], and may represent a key contributing factor in the process of atherosclerosis and its thrombotic complications. Oxidative stress has been identified as one of the factors closely associated with platelet hyperactivation in diabetes [12] and lipid hydroperoxides have been shown to stimulate platelet aggregation [15, 16]. In vitro, moreover, oxidised LDL [17, 18] and LDL from patients with type 1 [19] or type 2 diabetes [20] activated platelets via an increase of p38 mitogen-activated protein kinase (MAPK) phosphorylation and thromboxane B2 (TxB2) formation. To date, no data are available on how LDL from patients with the metabolic syndrome interact with platelets.

The objectives of the present study were: (1) to measure and compare different markers of oxidative stress in LDL from obese patients with the metabolic syndrome or type 2 diabetes, and from control volunteers; and (2) to determine the effects of these LDL on platelets.

Methods

Patients and control volunteers

We recruited 30 men (30–75 years old) from the Department of Endocrinology and Metabolic Diseases (Hospices Civils de Lyon). Exclusion criteria were smoking, alcohol consumption >2 units/day (20 ml), vitamin supplementation and evidence of cardiovascular complications. Patients with the metabolic syndrome were selected according to the criteria of the American Heart Association [21] and had to fulfil three or more of the following factors: waist circumference >102 cm, triacylglycerol >1.7 mmol/l, HDL-cholesterol <1 mmol/l, blood pressure >130/85 mmHg and fasting plasma glucose >5.6 mmol/l. Type 2 diabetic patients had to present with HbA1c greater than 7% (53 mmol/mol) and an associated feature of the metabolic syndrome. Patients treated with glitazone drugs were excluded and lipid-lowering drugs were suspended for at least 7 days before venepuncture. Control volunteers were in good health as assessed by medical history and had all above indices in the expected normal range. All participants gave written informed consent. The study protocol was approved by the Local Ethics Committee (CCP Sud Est IV, Hospices Civils de Lyon) and was in accordance with the principles of the Helsinki declaration.

Blood collection

Venous blood was collected on EDTA. Plasma was immediately separated by centrifugation at 1,500 g for 10 min and frozen at −80°C under nitrogen in the presence of 10 μmol/l butylhydroxytoluene.

Plasma secretory phospholipase A2 activity

Plasma secretory phospholipase A2 (sPLA2) activity was measured with a selective fluorescent substrate (Aterovax, Paris, France). Results are expressed in U/ml of sample, with 1 unit defined as the amount of sPLA2 enzyme that catalyses the release of 1 nmol of product in 1 min. The average intra-assay CV for individual human plasma samples was 8.46% and the average inter-assay CV was 5.24%. The minimum detectable activity level was 10 U/ml.

LDL isolation

LDL (1.019–1.063 g/ml) were isolated from plasma by potassium bromide stepwise ultracentrifugation (200,000 g) [22]. LDL preparations were extensively dialysed against PBS containing 1 mmol/l EDTA. LDL used to determine interactions with platelets were dialysed in a final step against EDTA-free PBS. The concentration of proteins was estimated using a Bradford assay [23]. Enzymatic determinations of total cholesterol and triacylglycerol were processed using commercial kits (Biomérieux, Marcy l’Etoile, France). Plasma high-sensitive C-reactive protein (CRP) was measured by nephelometry. LDL particle diameters were determined by electrophoresis of plasma in non-denaturating polyacrylamide gradient gels (Lara-Spiral, Couternon, France).

Fatty acid composition of lipid classes

Following the addition of internal standards and extraction with ethanol/chloroform (1:2, vol./vol.) in the presence of butylhydroxytoluene (50 μmol/l), lipid classes were separated by thin-layer chromatography. Chloroform/methanol (80:8, vol./vol.) was used to separate neutral lipids from phospholipids, and chloroform/methanol/water (63:27:4 by vol.) to separate phosphatidylcholines, phosphatidylethanolamines and sphingomyelins. Neutral lipids were separated into triacylglycerol and cholesteryl esters using hexane/diethylether/acetic acid (80:20:1 by vol.). The different spots were scraped off and treated for 90 min with trifluoride boron:methanol (1:1, vol./vol.) at 100°C. The derivatised fatty acid methyl esters and fatty dimethylacetals were extracted twice with isooctane and separated by gas chromatography using a gas chromatograph (HP 6890; Agilent Technologies, Massy, France) equipped with a capillary column (0.25 μm, 30 m × 0.25 mm, SP 2380; Supelco, Bellefonte, PA, USA) and a flame ionisation detector. The respective concentrations of fatty acids, quantified according to the amount of appropriate internal standard, were used to calculate the concentrations of lipid classes.

Quantification of total monohydroxylated fatty acids

Following lipid extraction, dried extracts were hydrolysed with 10 mol/l KOH for 20 min at 60°C [24]. Non-esterified hydroxylated fatty acids and fatty acids were first separated on a cartridge column (Oasis Sep-Pak; Waters, Milford, MA, USA) [3]. Non-esterified hydroxylated fatty acids were then separated by thin-layer chromatography with hexane/diethylether/acetic acid (60:40:1 by vol.), extracted with methanol, separated by reversed phase HPLC on an X Bridge C18 column (3.5 μm, 4.6 × 150 mm; Waters) using a gradient solvent of acetonitrile and water (pH 3), and detected at 235 nm (for details, see profile in electronic supplementary material [ESM] Fig. 1).

Malondialdehyde determination

LDL samples were mixed with thiobarbituric acid (10 mmol/l), acetic acid and butylhydroxytoluene (5 mmol/l), and incubated at 95°C for 60 min [25]. The thiobarbituric acid–malondialdehyde adducts were extracted with ethyl acetate, separated on to a Nucleosil C18 column (5 μm, 4.6 × 250 mm; Macherey-Nagel, Düren, Germany) and detected fluorimetrically (excitation 515 nm, emission 553 nm).

Vitamin E determination

LDL samples, containing tocol as an internal standard, were extracted twice with hexane (4 vol.) following the addition of ethanol (1 vol.) [20]. Tocopherol isomers were separated by reversed phase HPLC on to a Nucleosil C18 column (5 μm, 4 × 150 mm) and detected fluorimetrically (excitation 295 nm, emission 340 nm).

Platelet isolation and incubation with LDL

Blood was collected at the blood bank from healthy volunteers who had not ingested any aspirin or other non-steroidal anti-inflammatory drugs in the previous 10 days. Platelets were isolated [26] and incubated for 2 h at 37°C in the presence or absence of LDL (1 mg/ml).

Platelet p38 MAPK and cytosolic phospholipase A2 activation

Following platelet lysis, proteins (30 μg) were denatured, electrophoresed in 12% (wt/vol.) Bis-Tris gels (Bio-Rad Laboratories, Marnes-la-Coquette, France) and transferred to nitrocellulose membranes. The membranes were incubated with anti-p38 MAPK or antiphospho-p38 MAPK or anti-phospho-cytosolic phospholipase A2 (cPLA2) polyclonal antibodies (1:5,000 for all; Cell Signaling Technologies, Beverly, MA, USA), washed and then incubated with 1:10,000 goat anti-rabbit horseradish peroxidase conjugate. p38 MAPK, phospho-p38 MAPK and phospho-cPLA2 were visualised by enhanced chemiluminescence and bands were quantified by densitometry (Amersham Biosciences, Little Chalfont, UK).

TxB2 measurement

Platelet TxB2 was quantified by competitive enzyme immunoassay (Amersham Biosciences).

Platelet aggregation

Aggregation was measured in isolated platelets in a dual-channel aggregometer (Chrono-log; Coulter, Margency, France) according to the method of Born [27]. Platelet suspensions were pre-incubated for 5 min at 37°C in the presence or absence of LDL (0.1 mg/ml) and stimulated with sub-threshold concentrations of collagen (75 ± 9 ng/ml) while being continuously stirred at 1,000 rpm. The sub-threshold concentration of collagen was defined as the highest concentration of collagen that induced a less than 15% increase in light transmission. The extent of platelet aggregation was expressed in terms of percentage of change in light transmission at 4 min after the addition of collagen.

Statistical analysis

Results are expressed as the mean ± SEM. Comparisons between groups were performed using ANOVA followed by Fisher’s protected least significant difference (PLSD) post hoc test. Correlation coefficients were determined using linear regression analysis. With a statistical significance of p < 0.05 and 90% power, a minimum of eight volunteers per group was required to detect a 50% difference in the concentration of malondialdehyde in LDL.

Results

Clinical characteristics of obese patients with the metabolic syndrome or diabetes and of control volunteers

The clinical and metabolic characteristics of the participants are summarised in Table 1. Patients with the metabolic syndrome fulfilled the criteria of the American Heart Association [21]. They were overweight or obese, had abdominal obesity, mild hypertriacylglycerolaemia with low HDL-cholesterol, and were hypertensive. Of ten metabolic syndrome patients, two were being treated against hypertension, whereas four of the ten type 2 diabetic patients were taking anti-hypertensive medications. This could explain why mean systolic and diastolic blood pressure were higher in the metabolic syndrome than in the type 2 diabetes group. In addition, metabolic syndrome patients showed elevated fasting insulin concentrations and a tendency to have an increased HOMA index, both suggestive of insulin resistance. Levels of high-sensitivity CRP, a circulating marker of systemic inflammation, were increased to similar degrees in metabolic syndrome and type 2 diabetic patients. The type 2 diabetic patients were poorly controlled and also had features of the metabolic syndrome. LDL-cholesterol levels were lower in patients with the metabolic syndrome or type 2 diabetes than in control volunteers, while plasma triacylglycerol levels were higher. Healthy control volunteers had all above indices in the expected normal range.

Table 1 Clinical and metabolic characteristics of control volunteers and obese patients with the metabolic syndrome or type 2 diabetes
Full size table

sPLA2 activity was significantly increased in plasma from patients with the metabolic syndrome or type 2 diabetes (45.7 ± 4.3 and 51.9 ± 5.1 U/ml, respectively), compared with plasma from healthy volunteers (34.4 ± 2.5 U/ml). sPLA2 activity showed positive correlations with waist circumference (r 2 = 0.41, p < 0.001) and malondialdehyde concentrations in LDL (r 2 = 0.14, p < 0.05), and a negative correlation with plasmalogen proportions (r 2 = 0.15, p < 0.05).

Lipid and fatty acid composition of LDL from obese patients with the metabolic syndrome or diabetes, compared with that of LDL from control volunteers

As shown in Table 2, the concentrations of cholesteryl esters were lower in LDL from both patient groups (−13% in metabolic syndrome, −20% in type 2 diabetic patients) than in LDL from control volunteers. By contrast, concentrations of triacylglycerol were higher in LDL from metabolic syndrome (50%) and type 2 diabetic (72%) patients. No differences in glycerophospholipid concentrations (phosphatidylcholine and phosphatidylethanolamine) were observed between control and the two patient groups. LDL particle sizes were significantly lower in patients with the metabolic syndrome or type 2 diabetes than in control volunteers.

Table 2 Lipid classes and particle size of LDL from control volunteers and obese patients with the metabolic syndrome or type 2 diabetes
Full size table

The fatty acid composition of cholesteryl esters and phosphatidylcholine differed between patients and control volunteers (Table 3). The proportions of PUFA decreased in cholesteryl esters from metabolic syndrome patients, while saturated fatty acids (SFA) and mono-unsaturated fatty acids (MUFA) proportions increased. In particular, the proportion of linoleic acid (18:2 n-6), the main PUFA, decreased in cholesteryl esters and phosphatidylcholine from metabolic syndrome patients. Although palmitoleic acid (16:1 n-7) represents a minor fatty acid, its proportion increased by 70% in cholesteryl esters from that group. Concerning LDL from type 2 diabetic patients, PUFA proportions decreased while those of SFA increased and MUFA were unchanged. In particular, the proportions of linoleic acid decreased while those of arachidonic acid (20:4 n-6) increased in cholesteryl esters and phosphatidylcholine from type 2 diabetic LDL.

Table 3 Main fatty acid compositions of cholesteryl esters and phosphatidylcholines in LDL from control volunteers and obese patients with metabolic syndrome or type 2 diabetes
Full size table

Special care was taken to measure ethanolamine plasmalogens (or alkenyl,acyl-glycerophosphoethanolamines). They constitute a glycerophospholipid subclass with a 16:0, 18:0 or 18:1 fatty chain attached at the sn-1 position via a vinyl ether bond, and are more susceptible to oxidation [28]. As shown in Fig. 1a, the proportions of 16:0 dimethyl acetal issued from the alkenyl residue and of 18:0 dimethylacetal were decreased by 22% and 23%, respectively in LDL from metabolic syndrome patients compared with control LDL. In LDL from type 2 diabetic patients, 16:0, 18:0 and 18:1 dimethylacetal species were decreased by 47%, 50% and 42%, respectively. Altogether, the sum of dimethylacetal species, representative of plasmalogens, was decreased by 22% and 49% in LDL from patients with the metabolic syndrome or type 2 diabetes, respectively. Interestingly, ethanolamine plasmalogen proportions were negatively correlated with waist circumference (r 2 = 0.22, p < 0.01).

Fig. 1
figure 1

Indices of lipid peroxidation in LDL from control volunteers (white bars) and obese patients with metabolic syndrome (grey bars) or type 2 diabetes (black bars). a Fatty dimethylacetals (DMA) derived from phosphatidylethanolamine plasmalogens. b 9- and 13-HODE, (c) HODE/18:2 n-6 ratio and (d) malondialdehyde (MDA). Data are means ± SEM (n = 7–10 participants per group). p < 0.05 for 16:0 DMA, 18:0 DMA, DMA sum, 9-HODE, 13-HODE and HODE/18:2 n-6, and p < 0.01 for malondialdehyde in LDL from metabolic syndrome patients vs LDL from control volunteers; p < 0.05 for 9-HODE, 13-HODE and HODE/18:2 n-6, and p < 0.01 for 16:0 DMA, 18:0 DMA, 18:1 n-9 DMA, DMA sum and malondialdehyde in LDL from type 2 diabetic patients vs LDL from control volunteers; p < 0.05 for 16:0 DMA, 18:0 DMA, 18:1 n-9 DMA and DMA sum in LDL from type 2 diabetic patients vs LDL from metabolic syndrome patients. p values were obtained by ANOVA followed by Fisher’s PLSD post hoc test

Full size image

Lipid peroxides and antioxidant contents of LDL from obese patients with the metabolic syndrome or diabetes, compared with those of LDL from control volunteers

The concentrations of total (free and esterified) hydroxylated fatty acids, the stable primary products of lipid peroxidation, were determined in LDL from patient and control groups. They mainly consisted of hydroxy-octadecadienoic acids (HODE) derived from linoleic acid and HETE derived from arachidonic acid. Concentrations of 9-HODE and 13-HODE increased by 62% and 55% respectively in LDL from metabolic syndrome patients and by 52% and 50% in LDL from type 2 diabetic patients (Fig. 1b). Additionally, the ratio of total HODE to linoleic acid increased by 54% and 52% in LDL from metabolic syndrome and type 2 diabetic patients, respectively (Fig. 1c). Among HETE forms, 8-, 11-, 12- and 15-HETE were detected in LDL from each participant. There were no significant differences in the concentrations of HETE between metabolic syndrome patients and control volunteers (data not shown). Only the 15-HETE isomer concentration was significantly increased by 82% in LDL from type 2 diabetic patients, compared with LDL from control volunteers (50.8 ± 7.6 pmol/mg cholesterol vs 28.0 ± 4.3 pmol/mg cholesterol, p < 0.05). Concentrations of malondialdehyde, issued from the degradation of fatty acid hydroperoxides, were significantly increased by twofold in patients with the metabolic syndrome or type 2 diabetes (Fig. 1d). Waist circumference was mildly correlated with triacylglycerol and malondialdehyde levels in LDL (r 2 = 0.13 and 0.25, respectively, p < 0.05) (ESM Fig. 2a, b).

The concentrations of vitamin E, the main liposoluble antioxidant, were determined in LDL. α-Tocopherol concentrations did not differ between patients with the metabolic syndrome or type 2 diabetes (6.7 ± 0.5 and 6.0 ± 0.3 nmol/mg protein, respectively) vs control volunteers (5.9 ± 0.3 nmol/mg protein). γ-Tocopherol, which represents 10% of α-tocopherol, was increased in LDL from metabolic syndrome patients (0.9 ± 0.1 nmol/mg protein) compared with LDL from control (0.6 ± 0.1 nmol/mg protein), but did not differ in LDL from type 2 diabetic patients (0.7 ± 0.1 nmol/mg protein).

LDL isolated from obese patients with the metabolic syndrome or diabetes, or from control volunteers: effects on platelet arachidonic acid signalling cascade

To determine the effects of LDL on the platelet arachidonic acid cascade, platelets from healthy donors were incubated for 2 h at 37°C in the absence or presence of LDL from both patient groups or control volunteers (1 mg/ml). Phosphorylated amounts of p38 MAPK and cPLA2, key enzymes involved in the release of arachidonic acid from membrane phospholipids, as well as concentrations of TxB2, the stable catabolite of the pro-aggregatory arachidonic acid metabolite thromboxane A2, were determined. As shown in Fig. 2, the addition of LDL from control volunteers to platelets had no effects on p38 MAPK and cPLA2 phosphorylation, compared with platelets alone. By contrast, the addition of LDL from patients with the metabolic syndrome or type 2 diabetes significantly increased p38 MAPK phosphorylation, by 21% and 55% respectively, and cPLA2 phosphorylation by 23% and 24% respectively. Compared with control platelets, the incubation of platelets with LDL from patients with the metabolic syndrome or type 2 diabetes resulted in a twofold increased basal concentration of TxB2, whereas LDL from control volunteers had no effects (Fig. 3). Platelet TxB2 was negatively correlated with plasmalogen proportions (r 2 = 0.44) and positively with HODE and malondialdehyde concentrations in LDL (r 2 = 0.24 and r 2 = 0.31, respectively, p < 0.01) (ESM Fig. 2c, d).

Fig. 2
figure 2

Effects of LDL from control volunteers and obese patients with metabolic syndrome or type 2 diabetes on the phosphorylation of (a) p38 MAPK and (b) cPLA2 in platelets. Platelets (Pl) from blood donors were incubated for 2 h at 37°C in the absence or presence of LDL (1 mg protein/ml) from either control volunteers (C), metabolic syndrome patients (MS) or type 2 diabetic patients (T2D). Representative immunoblots and histograms of the normalised amount of phosphorylation of p38 MAPK and cPLA2 are shown. Data are means ± SEM (n = 7–8 participants per group). p < 0.05 for phospho-p38 MAPK and phospho-cPLA2 in Pl+MS vs Pl; p < 0.05 for phospho-cPLA2 and p < 0.01 for phospho-p38 MAPK in Pl+T2D vs Pl; p < 0.05 for phospho-p38 MAPK in Pl+T2D vs Pl+MS. p values were obtained by ANOVA followed by Fisher’s PLSD post hoc test

Full size image
Fig. 3
figure 3

Effects of LDL from control volunteers and obese patients with metabolic syndrome or type 2 diabetes on TxB2 concentrations in unstimulated platelets. Platelets (Pl) from blood donors were incubated for 2 h at 37°C in the absence or presence of LDL (1 mg protein/ml) from either control volunteers (C), metabolic syndrome patients (MS) or type 2 diabetic patients (T2D). TxB2 concentrations in control unstimulated platelets were 187 ± 37 pmol/109 platelets. Data, expressed as per cent of control, are means ± SEM (n = 10 participants per group). p < 0.05 for Pl+T2D vs Pl and Pl+MS vs Pl. p values were obtained by ANOVA followed by Fisher’s PLSD post hoc test

Full size image

Effect on platelet aggregation of LDL isolated from obese patients with the metabolic syndrome or diabetes, or from control volunteers

To determine the effect of LDL on platelet aggregation, platelets were pre-incubated for 5 min with low concentrations of LDL (0.1 mg/ml) and then stimulated with sub-threshold concentrations of collagen for another 4 min. As shown in Fig. 4, LDL from control volunteers had no effect on collagen-induced platelet aggregation, compared with platelets incubated with buffer. By contrast, pre-incubation of platelets with LDL from metabolic syndrome or type 2 diabetic patients resulted in stimulation of platelet aggregation in response to sub-threshold concentrations of collagen (197% and 251%, respectively). Platelet aggregation was also potentiated by lower concentrations (50 μg/ml) of LDL from type 2 diabetic patients (data not shown).

Fig. 4
figure 4

Effects of LDL from control participants and obese patients with metabolic syndrome or type 2 diabetes on the aggregation of platelets induced by collagen. Platelets (Pl) from blood donors were pre-incubated for 5 min at 37°C in the absence or presence of LDL (0.1 mg protein/ml) from either control volunteers (C), metabolic syndrome patients (MS) or type 2 diabetic patients (T2D) and stimulated with sub-threshold concentrations of collagen (75 ± 9 ng/ml). Data, expressed as per cent of aggregation, are means ± SEM of five participants per group. p < 0.05 for Pl+MS vs Pl; p < 0.01 for Pl+T2D vs Pl. p values were obtained by ANOVA followed by Fisher’s PLSD post hoc test

Full size image

Discussion

The evidence for increased oxidative stress in patients with the metabolic syndrome is still a matter of debate. In the present study, we provide several lines of evidence for increased oxidative stress in LDL from obese metabolic syndrome patients, as assessed by different complementary indices of oxidative stress and by special focus on oxidation products of linoleic acid, the most abundant PUFA in LDL. Higher concentrations of total HODE and lower proportions of linoleic acid in phosphatidylcholine and cholesteryl esters were found in LDL from metabolic syndrome patients compared with control LDL. Interestingly, HODE and linoleic acid levels were modified to the same extent in LDL from type 2 diabetic patients. Conversely, there were no changes in total HETE concentrations in metabolic syndrome patients, which is in agreement with a recent study showing no differences in the levels of oxidation products of arachidonic acid (total HETE and F2-isoprostanes) in individuals with features of metabolic syndrome [9]. This suggests that oxidation products of linoleic acid could be better markers of lipid peroxidation than those of arachidonic acid. It should be mentioned that participants did not supply a dietary record. Although we are not aware of any supplementation study reporting a concomitant decrease of linoleic acid and increase of arachidonic acid in plasma or LDL, we cannot exclude a potential effect of fat intake on plasma fatty acid composition.

The presence of oxidative stress in LDL from metabolic syndrome patients was confirmed by the decreased concentrations of plasmalogens in LDL from metabolic syndrome patients (−22%) as well as in LDL from type 2 diabetic patients (−49%), compared with control LDL. This result indicates that plasmalogens may be targets and markers of ongoing oxidative stress, and corroborates previous data showing a 20% decrease of plasmalogen phosphatidylethanolamine in erythrocyte membranes from hyperlipidaemic patients [29]. In addition, we found that malondialdehyde, a commonly used overall marker of lipid peroxidation that has been described to correlate with total isoprostanes in plasma and tissues [30, 31], was increased to a similar extent in LDL from metabolic syndrome and type 2 diabetic patients, in agreement with the enhanced serum levels of thiobarbituric acid reactive species observed in type 2 diabetic patients [32]. Although isoprostanes are recognised as reliable markers of oxidative stress in vivo, they derive exclusively from the free radical-induced peroxidation of arachidonic acid and their levels were either reported to be increased [7] or unchanged [9] in patients with the metabolic syndrome. Altogether, the increase of several complementary indices of oxidative stress and especially of oxidised products of linoleic acid in LDL from metabolic syndrome patients supports the hypothesis that the metabolic syndrome is associated with higher levels of circulating oxidised LDL [4–6]. In addition, compared with control LDL, LDL from patients with the metabolic syndrome showed higher triacylglycerol and lower cholesteryl ester levels as well as decreased particle size, suggesting a higher percentage of smaller and denser LDL particles. Therefore, our data corroborate reports that the metabolic syndrome is associated with small and dense LDL particles [5, 33, 34], which are known to be more susceptible to oxidative modification [35].

Regarding the main antioxidant in LDL, we found similar concentrations of α-tocopherol in patients with the metabolic syndrome or type 2 diabetes, and higher concentrations of γ-tocopherol in metabolic syndrome patients. There have been discordant reports about vitamin E levels in patients with metabolic syndrome or type 2 diabetes, showing either lower vitamin Ε concentrations in participants who had a feature of the metabolic syndrome and also consumed fewer fruits and vegetables [36], or no difference among participants with the metabolic syndrome or type 2 diabetes [37, 38]. Only one study to our knowledge has reported slightly higher serum γ-tocopherol levels [38] and no changes of α-tocopherol among participants with the metabolic syndrome. Our results obtained ex vivo suggest that tocopherols may have been spared by molecules (such as plasmalogens) serving as primary targets of free radicals during LDL oxidation [39].

It is worth noting that oxidative stress was enhanced to a similar extent in LDL from metabolic syndrome and from type 2 diabetic patients, suggesting that factors associated with the metabolic syndrome, rather than poor glycaemic control, are key factors associated with oxidative modification of LDL. In agreement with this statement, our results show correlations between waist circumference and plasmalogens or malondialdehyde in LDL, suggesting an association between an excess of visceral fat and biochemical variables associated with oxidative stress in LDL. It is also likely that obesity tightly associated with the metabolic syndrome may represent a major factor in the oxidative stress seen in LDL from obese metabolic syndrome patients.

Beside the above, our results show that the activity of sPLA2 is correlated with plasmalogens and malondialdehyde in LDL, suggesting that this enzyme catalyses the hydrolysis of oxidised phospholipids in LDL [40]. The correlation between the activity of sPLA2 and waist circumference is in line with the association of sPLA2 with the metabolic syndrome [41].

Regarding the functionality of LDL from patients with metabolic syndrome, we show for the first time that LDL from metabolic syndrome patients activate platelets and prime collagen-induced platelet aggregation. The activation occurs through an increased phosphorylation of p38 MAPK, the stress kinase responsible for the phosphorylation of cPLA2 [42], the key enzyme involved in the release of arachidonic acid from membrane phospholipids [43]. It resulted in increased formation of thromboxane A2, as assessed by the increased concentration of the stable catabolite, TxB2. We confirm that LDL isolated from type 2 diabetic patients activate platelets through the arachidonic acid cascade [20] and give new evidence for a priming effect of LDL on platelet aggregation, whereas LDL from control volunteers had no activating effects. As for the question of which molecules in LDL from the two patient groups are responsible for platelet activation, our data suggest that lipid peroxides may be implicated. LDL from obese patients with metabolic syndrome or type 2 diabetes showed similarly increased levels of lipid peroxides and activated platelets. In addition, correlation analyses show that the platelet TxB2 concentration is related to concentrations of HODE and malondialdehyde in LDL, suggesting that lipid peroxides could underlie the activating effects of LDL on platelets and that glycation is not necessarily required. In future studies it will be important to take into account obesity in order to identify the mechanism by which oxidised LDL activate platelets.

In addition, our results obtained ex vivo strengthen the notion that fatty acid peroxides are implicated in platelet activation as previously demonstrated in in vitro studies [15, 44]. Indeed, we have previously shown that fatty acid hydroperoxides, precursors of HETE and HODE, potentiated platelet aggregation and activated platelets via the arachidonic acid signalling cascade [16]. More recently, the Podrez group showed that oxidised choline glycerophospholipids induce platelet aggregation via CD36 [45].

Altogether, our results support the idea that: (1) oxidative stress is increased in LDL from obese metabolic syndrome patients with or without type 2 diabetes; and (2) in vivo oxidatively modified LDL from metabolic syndrome patients activate control platelets to the same extent as LDL from type 2 diabetic patients. Considering that the metabolic syndrome often precedes the occurrence of diabetes [46] and that cardiovascular complications occur frequently at the onset of type 2 diabetes, our data suggest that oxidative stress in metabolic syndrome is an early promoter of premature atherothrombosis [47]. The identification of specific targets of oxidative stress in the metabolic syndrome, such as linoleic acid-containing lipids, might provide opportunities to selectively slow down some specific lipid peroxidation processes in LDL at the stage of the metabolic syndrome and to prevent the associated platelet hyperactivation.