- Email: [email protected]
Plasma levels of soluble CD36, platelet activation, inflammation, and oxidative stress are increased in type 2 diabetic patients
Free Radical Biology & Medicine 52 (2012) 1318–1324
Contents lists available at SciVerse ScienceDirect
Free Radical Biology & Medicine journal homepage: www.elsevier.com/locate/freeradbiomed
Original Contribution
Plasma levels of soluble CD36, platelet activation, inflammation, and oxidative stress are increased in type 2 diabetic patients Rossella Liani a, Bente Halvorsen b, Simona Sestili a, Aase Handberg c, Francesca Santilli a, Natale Vazzana a, Gloria Formoso a, Pål Aukrust b, Giovanni Davì a,⁎ a b c
Center of Excellence on Aging, University of Chieti “G. d'Annunzio,” Via Colle dell'Ara, 66013, Chieti, Italy Research Institute of Internal Medicine, Oslo University Hospital Rikshospitalet, University of Oslo, Oslo, Norway Department of Clinical Biochemistry, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark
a r t i c l e
i n f o
Article history: Received 10 August 2011 Revised 31 January 2012 Accepted 5 February 2012 Available online 15 February 2012 Keywords: sCD36 Type 2 diabetes Platelet activation Oxidative stress
a b s t r a c t Inflammation, oxidative stress, and platelet activation are involved in type 2 diabetes and its complications. Soluble CD36 (sCD36) has been proposed to early identify diabetics at risk of accelerated atherothrombosis. We aimed at characterizing the platelet contribution to sCD36 in diabetes, by correlating its concentration with the extent of platelet-mediated inflammation and in vivo lipid peroxidation and investigating the effects of low-dose aspirin on these processes. A cross-sectional comparison of sCD36, soluble CD40L (sCD40L) reflecting platelet-mediated inflammation, urinary 11-dehydro-TxB2, and 8-iso-PGF2α, in vivo markers of platelet activation and lipid peroxidation, was performed among 200 diabetic patients (94 of them on aspirin 100 mg/day) and 47 healthy controls. sCD36 levels (median [IQR]: 0.72 [0.31–1.47] vs 0.26 [0.2–0.37], P = 0.003) and urinary 11-dehydro-TxB2 levels (666 [293–1336] vs 279 [160–396], P ≤ 0.0001) were significantly higher in diabetic patients not on aspirin (n = 106) than in healthy subjects. These variables were significantly lower in aspirin-treated diabetics than untreated patients (P b 0.0001). Among patients not on aspirin, those with long-standing diabetes (> 1 year) had significantly higher sCD36 levels in comparison to patients with diabetes duration b 1 year (1.01 [0.62–1.86] vs 0.44 [0.22–1.21], P = 0.001). sCD36 linearly correlated with sCD40L (rho = 0.447; P = 0.0001). On multiple regression analysis, 11-dehydro-TxB2 (β = 0.360; SEM = 0.0001, P = 0.001), 8-iso-PGF2α (β = 0.469; SEM = 0.0001, P b 0.0001), and diabetes duration (β = 0.244; SEM = 0.207, P = 0.017) independently predicted sCD36 levels. sCD36, platelet activation, inflammation, and oxidative stress are increased in type 2 diabetes. Future studies are needed to elucidate if the incomplete down-regulation of sCD36 by low-dose aspirin implies that sCD36 may be derived from tissues other than platelets or if additional antiplatelet strategies in diabetes are necessary to interrupt CD36dependent platelet activation. © 2012 Elsevier Inc. All rights reserved.
Introduction Atherogenesis starts when low-density lipoprotein (LDL) accumulates in the vessel wall at sites of injury and is oxidized to ox-LDL, accumulating into macrophages. The resulting foam cells are characteristic for the early atherosclerotic lesions [1]. The scavenger receptor CD36 is believed to have a critical role in the initiation and progression of atherosclerosis through its ability to bind and internalize modified LDL, facilitating the formation of macrophage foam cells [2].
Abbreviations: 11-dehydro-TxB2, 11-dehydro-thromboxane B2; 8-iso-PG2α, 8-isoprostaglandin F2-alpha; ASA, acetylsalicylic acid; BMI, body mass index; CRP, C-reactive protein; HbA1c, hemoglobin A1c; IL, interleukin; LDL, low-density lipoprotein; ox, oxidized; PCOS, polycystic ovary syndrome; ROS, reactive oxygen species; sCD36, soluble CD36; sCD40L, soluble CD40 ligand; T2DM, type 2 diabetes mellitus. ⁎ Corresponding author. Fax: + 39 0871 541261. E-mail address: [email protected] (G. Davì). 0891-5849/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2012.02.012
CD36 is an 88-kDa glycosylated transmembrane protein expressed in various cell types, including macrophages and endothelial cells [3]. CD36 is involved in physiological and pathological processes in vivo, including atherosclerosis and diabetes [3]. Platelets express scavenger receptors, including CD36 [4], although their functional role in platelet biology has not been clearly defined. A specific CD36-dependent signaling pathway is required for platelet activation by ox-LDL [5]. Ox-LDL is present in the circulation, and platelets can come into contact with ox-LDL and become activated, thereby contributing to the thrombotic process [6]. Platelet CD36 is associated with nonreceptor tyrosine kinases of the Src family [7], implicated in platelet activation by ox-LDL [8]. CD36 engagement may induce an activating signal that is synergistic with signals from other receptors and may result in platelet activation by subthreshold concentrations of physiological agonists. Thus, CD36 joins a growing number of platelet receptors and ligands that may potentiate members of the platelet activation cascade, including Gas6 and its
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
receptors, CD40 ligand, αIIbβ3, Eph kinases and ephrins. Unlike other “coreceptor” ligands, which are primarily localized at the platelet surface during the initial phase of platelet aggregation, CD36 ligands are likely to be presented to the platelet surface before “classical” platelet agonists. Thus, CD36 may serve to “prime” or sensitize the platelet for subsequent activation, and this may contribute to platelet hyperreactivity [9]. Besides ox-LDL, urinary F2-isoprostanes levels represent reliable and sensitive markers of in vivo ongoing lipid peroxidation [10]. Isoprostanes are a family of bioactive compounds produced from arachidonic acid via a free radical-catalyzed mechanism of lipid peroxidation on cell membrane phospholipids or circulating LDL [11]. Moreover, F2-isoprostanes, in particular 8-iso-PGF2α, may amplify the aggregation response to subthreshold concentrations of platelet agonists [10]. This might be relevant to settings in which platelet activation and enhanced free radical formation coincide, such as diabetes mellitus [12]. 8-iso-PGF2α is formed during the in vitro oxidation of LDL [13] and circulating ox-LDL is closely related to isoprostane levels [14]. It has been previously demonstrated that a soluble form of CD36 (sCD36) is present in plasma [15]. In type 2 diabetic patients (T2DM), plasma CD36 is markedly elevated compared to control subjects [15], and it is highly related to insulin resistance and glycemic control. It has been proposed that sCD36 is derived from the surfaces of monocytes and macrophages with increased CD36 expression [15]. Indeed, patients with T2DM have been reported to have enhanced monocyte CD36 expression, potentially reflecting increased posttranscriptional efficiency of CD36 mRNA in response to chronic hyperglycemia [16]. However, platelet activation is a common feature of T2DM [17], and platelets could therefore also contribute to the enhanced levels of sCD36 in these patients. In the present study, we tested the hypothesis that the increased concentrations of sCD36 observed in T2DM may derive, at least in part, from platelet activation. Therefore, the goals of this study were to characterize the platelet contribution to sCD36 in T2DM, by correlating its concentration with the extent of platelet-mediated inflammation and in vivo lipid peroxidation as reflected by enhanced isoprostane formation, and investigating the effects of low-dose aspirin on these processes.
1319
the American Diabetes Association [18,19] and 47 healthy subjects (31 female, 16 male; aged 39 ± 12.8 years) were enrolled in the study. The baseline characteristics of diabetic patients are reported in Table 1. Exclusion criteria were: (1) a recent history (b6 months) of thrombotic events, pregnancy, or lactation; (2) regular use of estroprogestin, iron, antioxidants (vitamins C and E), NSAIDs, antiplatelet agents other than aspirin; (3) clinically significant hepatic, renal, cardiac, or pulmonary insufficiency; (4) history of malignant neoplasms (diagnosed and treated within the last 5 years); (5) autoimmune disorders and type 1 diabetes mellitus. At the time of the study, diabetic patients were being treated by diet alone, insulin alone, diet plus hypoglycemic agent, or hypoglycemic agent plus insulin. Among them, 94 diabetic patients were also being treated with low-dose aspirin (100 mg/day) for primary or secondary cardiovascular prevention [19]. Diabetic patients were examined for the presence of microvascular and macrovascular complications. Of the 200 patients, 27 (13.5%) (20 on ASA treatment) had microvascular complications (nephropathy, detected by the presence of persistent microalbuminuria between 30 and 300 mg/24 h, in at least two of three consecutive 24-h collections, with a eGRF ≥ 90 ml/min/1.73 m 2, as calculated on MDRD formula; retinopathy, based on fundus oculi examination) and 38 (19%) (35 on ASA treatment) had a history or physical examination positive for evidence of macrovascular complications (cerebrovascular disease, coronary heart disease, or peripheral arterial disease). Reasons for lack of aspirin treatment included unwillingness of the patient (one patient) or recent bleeding episodes, eliciting temporary drug withdrawal (2 patients). Diabetic patients with arterial hypertension or hypercholesterolemia were included if well controlled with stable drug therapy: 96 (48%) had arterial hypertension (57 on ASA treatment), defined as current systolic/diastolic blood pressure >130/85 mm Hg; 131 (65.5%) were hypercholesterolemic (71 on ASA treatment), in accordance with the ATPIII criteria [20]. Written informed content was obtained from each subject participating in the study. The local Ethics Committee approved the protocol.
Design of the studies Materials and methods Subjects Two-hundred patients (91 female, 109 male; aged 64.1 ± 8.7 years) with T2DM, as defined in accordance with the criteria of
First, a cross-sectional comparison of circulating sCD36, soluble CD40L (sCD40L) as a marker of platelet-mediated inflammation, urinary 11-dehydro-TxB2, and 8-iso-PGF2α was performed among all patients and controls. All subjects were studied as outpatients after a 12-h fast and performed an overnight urine collection immediately
Table 1 Baseline characteristics of the type 2 diabetic patients. Type 2 diabetes mellitus (n = 200) Variables
Male gender Age (years) BMI (kg/m2) Diabetes duration b 1 year > 1 year Fasting plasma glucose (mmol/L) Hemoglobin A1c (%) Hypertension Hypercholesterolemia Microvascular complications Macrovascular complications
All (n = 200)
Not on ASA (n = 106)
On ASA (n = 94)
n (%) Median Median [IQR]
109 (54.5) 64 [60 to 69.7] 28.1 [25.1 to 31.1]
52 (49) 64 [57.5 to 70] 28.6 [25.4 to 32]
57 (60.6) 66 [61 to 69] 28.1 [24.9 to 30.8]
n (%)
85 (42.5) 115 (57.5) 7.4 [6.6 to 8.5] 6.9 [6.4 to 7.5] 96 (48) 131 (65.5) 27 (13.5) 38 (19)
63 (59.4) 43 (40.6) 7.7 [6.8 to 9.5] 6.9 [6.4 to 7.3] 39 (36.8) 60 (56.6) 7 (6.6) 3 (2.8)
22 (23.4) 72 (76.6) 7.1 [6.2 to 8.3] 6.9 [6.4 to 7.6] 57 (60.6) 71 (75.5) 20 (21.3) 35 (37.2)
Median [IQR] Median [IQR] n (%) n (%) n (%) n (%)
P values 0.118 0.051 0.233 b 0.0001 0.008 0.552 0.001 0.843 0.003 b 0.0001
1320
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
before blood sampling. Urine samples were added with the antioxidant 4-hydroxy-Tempo (1 mM) (Sigma Chemical Co., St. Louis, MO) and stored at -20 °C until extraction. Blood samples were obtained for lipid levels as well as routine blood chemistry to check inclusion and exclusion criteria. To assess the potential influence of long-standing glucose control impairment on sCD36 levels, in the 106 T2DM patients not on ASA treatment, sCD36 levels were also compared between subjects who had a known diabetes duration b1 year, and subjects with longstanding diabetes (>1 year). To test the hypothesis of a platelet origin of sCD36, 94 T2DM patients on aspirin (100 mg/day) were also studied. Finally, we tested the hypothesis of an association among CD36 release in plasma, oxidative stress, and platelet activation in 16 T2DM patients, not on ASA treatment, in whom measurements were repeated after 18–20 weeks. Routine biochemical measurements Fasting plasma glucose was measured by the glucose oxidase method. The HbA1c level was determined by automated highperformance liquid chromatography (HPLC) [21]. Enzyme-linked immunosorbent assay (ELISA) The sCD40L was determined by ELISA (R&D Systems, Minneapolis, MN). The sCD36 was measured using an in-house ELISA [15]. A pool of EDTA plasma, aliquoted and stored at − 80 °C, was applied in seven dilutions in each run and used as a standard concentration curve. Two dilutions of another EDTA pool were used as internal controls, and each control was run in quadruplicate on each ELISA plate. Patient EDTA plasma samples, which were stored at −80 °C, were analyzed in duplicate. Runs were accepted if the controls were within ± 2 SD from mean, and most were within 1 SD. Intraassay CV was 6%, and day-day-assay CV was 16.4%, estimated from internal controls in the study. Standard curves were logtransformed and thereby linear. A few measurements outside the range of the standard curve were calculated from the extrapolated standard curve. In the text, the concentration of sCD36 is given in relative units.
variables significantly related to plasma CD36 in the univariate analysis as covariates. Data are presented as mean (± standard deviation) or as median and interquartile range (IQR) (25th, 75th percentile). Only P values b0.05 were regarded as statistically significant. All tests were twotailed, and analyses were performed using a computer software package (or Statistical Package for the Social Sciences, version 18.0, SPSS Inc., Chicago, IL).
Results Plasma levels of CD36 in T2DM and healthy controls in relation to ASA therapy Soluble CD36 levels were significantly higher in patients with T2DM not on ASA treatment than in healthy subjects (median [IQR]: 0.72 [0.31 to 1.47] vs 0.26 [0.2 to 0.37], P = 0.003), whereas plasma CD36 levels were significantly lower in diabetic patients on ASA treatment when compared with diabetic patients not on ASA (0.39 [0.27 to 0.58] vs 0.72 [0.31 to 1.47], P b 0.0001) but still significantly higher than healthy subjects (0.39 [0.27 to 0.58] vs 0.26 [0.2 to 0.37], P b 0.0001) (Fig. 1A). T2DM patients not on ASA treatment had a significantly higher urinary 11-dehydro-TxB2 excretion rate in comparison with control subjects (666 [293 to 1336] vs 279.5 [160 to 396.5] pg/mg creatinine, P ≤ 0.0001). Urinary 11-dehydro-TxB2 levels were significantly lower in diabetic patients treated with ASA, in comparison with nonaspirin-treated subjects, including diabetic patients not on ASA treatment (146 [100.5 to 247] vs 666 [293 to 1336] pg/mg creatinine, P ≤ 0.0001) and healthy subjects (146 [100.5 to 247] vs 279.5 [160 to 396.5] pg/mg creatinine, P b 0.0001) (Fig. 1B).
Urinary eicosanoid assays Urinary 8-iso-PGF2α and 11-dehydro-TxB2 were measured by previously described radioimmunoassay methods that have been validated by using different antisera and by comparison with gas chromatography/mass spectrometry [22,23]. Statistical analysis Sample size calculation was based on previous data reporting a 4.5-fold difference in plasma CD36 concentration between obese diabetic subjects and lean controls [15]. With 106 patients not on aspirin and 47 controls, the study had a 99% power to detect a difference of at least 3.5-fold between diabetic patients and controls. The Kolmogorov-Smirnov test was used to determine whether each variable had a normal distribution. When necessary, log transformation was used to normalize the data, or appropriate nonparametric tests were used (Mann-Whitney U test; Spearman correlation coefficient). Differences at baseline in categorical variables were analyzed by Fisher's exact test. Comparisons between groups were made with ANOVA with the Scheffè post hoc test to correct for multiple comparisons, with logarithmically transformed data. Comparisons within groups were made with the Wilcoxon test. A multiple linear regression analysis was performed to further quantify the relationship between the sCD36 and the variables in study, with
Fig. 1. Plasma levels of CD36 in T2DM and healthy controls in relation to ASA therapy. Plasma levels of CD36 (A) and urinary 11-dehydro-TxB2 (B) excretion rate in healthy subjects, and in diabetic patients on aspirin and not on aspirin. Data are median (interquartile range). Tx = thromboxane.
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
1321
Plasma levels of CD36 in relation to oxidative stress and markers of platelet activation A significant direct correlation was found between plasma CD36 and urinary 11-dehydro-TxB2 and 8-iso-PGF2α excretion rates (rho = 0.393; P = 0.0003, and rho = 0.481; P = 0.0001, respectively) in diabetic patients not on ASA treatment (Figs. 3A and B). Moreover, a significant linear correlation was found between plasma CD36 and CD40L (rho = 0.447; P = 0.0001) (Fig. 3C). Plasma levels of CD36 in T2DM in relation to metabolic variables, atherosclerotic risk factors, and prostanoid metabolites—Multivariate analyses To further define the relationship between sCD36, metabolic variables, additional atherosclerotic risk factors and urinary prostanoid metabolites in the 106 T2DM not treated with ASA, a multiple regression analysis was performed at baseline in which sCD36 was included
Fig. 2. Plasma levels of CD36 in relation to diabetes duration. Comparison of soluble CD36 in the 106 patients not on ASA treatment, subdivided in two groups: diabetes duration b 1 year and long-standing diabetes (> 1 year). Correlation between fasting plasma glucose and sCD36 (rho = 0.378; p = 0.002) in diabetic patients not on aspirin. Correlation between hemoglobinA1c and sCD36 (rho = 0.255; p = 0.026) in diabetic patients not on aspirin.
Plasma levels of CD36 in relation to diabetes duration T2DM patients not treated with ASA and with long-standing diabetes (>1 year) had significantly higher plasma CD36 levels in comparison with T2DM patients with diabetes duration b1 year (1.01 [0.62 to 1.86] vs 0.44 [0.22 to 1.21], P = 0.001) (Fig. 2A). Consistently, a significant linear correlation was found between plasma CD36 and diabetes duration (rho = 0.347; P = 0.0001). Furthermore in the same group of patients, a significant linear correlation was found between plasma CD36 and fasting blood glucose and HbA1c (rho = 0.378; P = 0.002, and rho = 0.255; P = 0.026, respectively) (Figs. 2B and C). On the contrary, in diabetic patients on ASA, no significant correlation was found between plasma CD36 and HbA1c (rho = 0.050; P = 0.648). No significant relationship was found, either in diabetic patients on ASA or not on ASA, between plasma creatinine and urinary 11dehydro-TxB2 and 8-iso-PGF2α (on ASA subjects rho = -0.127, P = 0.359 and rho = -0.065, P = 0.641, respectively; not on ASA subjects rho = 0.020, P = 0.924 and rho = -0.138, P = 0.520, respectively).
Fig. 3. Plasma levels of CD36 in relation to oxidative stress and markers of plateletactivation. Correlation between urinary 11-dehydro-TxB2 and sCD36 (rho = 0.393; p = 0.0003) in the diabetic patients not on aspirin. Correlation between urinary 8iso-PGF2α and sCD36 (rho = 0.481; p = 0.0001) in diabetic patients not on aspirin. Correlation between plasma levels of CD40L and sCD36 (rho = 0.447; p = 0.0001) in diabetic patients not on aspirin. Tx = thromboxane; PG = prostaglandin.
1322
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
sCD36. In contrast, only 32.3% of T2DM patients with 11-dehydroTxB2 and 8-iso-PGF2α urinary excretion in the first and second quartiles had increased levels of sCD36 (P = 0.004). Plasma levels of CD36 during longitudinal testing The reproducibility of plasma CD36 and CD40L measurements as well as thromboxane biosynthesis and isoprostane formation was assessed by obtaining an additional sample 18-20 weeks later, from 16 patients with T2DM not on ASA treatment. No significant differences in plasma CD36, and CD40L, urinary 11-dehydro-TxB2, and 8iso-PGF2α assessed on the two different occasions (from 0.42 [0.26 to 0.69] to 0.39 [0.22 to 0.56], P = 0.485; from 0.5 [0.16 to 1.06] to 0.42 [0.17 to 1.23] ng/mL, P = 0.570; from 1095 [680.75 to 2052.75] to 1085 [756.51 to 2266.5] pg/mg creatinine, P = 0.245; from 410 [243.75 to 584] to 450.5 [284.75 to 600.64] pg/mg creatinine, P = 0.196, respectively) were detected (Fig. 5A, B, C and D). No detectable change in metabolic control or body weight occurred throughout the 18–20-week period between assessments. Fig. 4. Plasma levels of CD36 in T2DM in relation to metabolic variables, atherosclerotic risk factors and prostanoid metabolites – multivariate analyses. Correlation between 8iso-PGF2α and 11-dehydro-TxB2 excretion rates in type 2 diabetic patients according to quartiles of plasma CD36. Vertical and horizontal lines mark the boundaries of median individual measurements; open circles = first and second quartile of sCD36; solid circles = third and fourth quartile of sCD36. Data are presented on a logarithmic scale. PG = prostaglandin; Tx = thromboxane.
as the dependent variable. Stepwise linear regression yielded a model in which only urinary 11-dehydro-TxB2 (regression coefficient = 0.360; standard error of the mean = 0.0001, P = 0.001) and 8-iso-PGF2α (regression coefficient = 0.469; standard error of the mean = 0.0001, P b 0.0001) and diabetes duration (regression coefficient = 0.244; standard error of the mean = 0.207, P = 0.017) predicted sCD36 levels, independently of gender, fasting blood glucose, and hypercholesterolemia. Thus, as shown in Fig. 4, among 36 diabetic patients with 11-dehydro-TxB2 and 8-iso-PGF2α excretion rates in the third and fourth quartiles, 25 (69.4%) had increased levels of
Discussion Soluble CD36 (sCD36) has been proposed to reflect monocyte/ macrophage and platelet CD36 expression in the metabolic syndrome, with a consistent relationship between insulin resistance and sCD36, potentially being a marker that integrates insulin resistance and atherosclerosis [15]. In fact circulating CD36 may be present as the unbound protein or a peptide fraction thereof or may be present in microparticles shed from cells such as platelets, monocytes/macrophages, or adipocytes after being triggered by various stimuli. Further dedicated studies are required to support this hypothesis [15]. Increased plasma CD36 levels have been described not only in overt T2DM [24] but also in prediabetic conditions such as polycystic ovary syndrome (PCOS) and in subjects with impaired glucose tolerance [25]. sCD36 levels seem to be predicted by chronic low-grade inflammatory state (C-reactive protein [CRP] and interleukin [IL]-6), insulin sensitivity, and BMI [25,26], all involved in the
Fig. 5. Plasma levels of CD36 during longitudinal testing. Urinary 8-iso-PGF2α (A), soluble CD36 (B), urinary 11-dehydro-TxB2 (C), and soluble CD40L (D) in 16 diabetic patients not on aspirin. Individual patient data are also represented in a second sample obtained 18-20 weeks later. Tx = thromboxane; PG = prostaglandin.
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
pathogenesis of diabetes and metabolic syndrome. Thus, sCD36 has been proposed as a simple marker to predict, at an early stage, diabetic subjects at risk of developing accelerated atherothrombosis. Platelets express CD36 [4], and a specific CD36-dependent signaling pathway is required for platelet activation by ox-LDL [5]. Moreover, CD36 may “prime” or sensitize the platelet for subsequent activation, contributing to platelet hyperreactivity [9]. However, the contribution of persistent platelet activation to increased shedding of CD36 has not been previously investigated. T2DM is characterized by persistent thromboxane-dependent platelet activation [17,27]. In the present study we demonstrated that T2DM patients have persistently increased levels of sCD36, related to the impaired metabolic control (plasma glucose levels and HbA1c) and to diabetes duration. Moreover, the highly significant correlation between plasma sCD36 levels and the urinary excretion rate of 11-dehydro-TxB2, a noninvasive index of in vivo platelet activation [27], supports the likelihood of CD36 release during TxA2-dependent platelet activation in T2DM. This relationship is strengthened by the correlation between plasma CD40L and sCD36, considering that activated platelets are the major cellular source of sCD40L in the circulation [28], and that platelet CD40L itself provides a mechanism for platelet activation [29]. Both these positive correlations are in keeping with previously reported evidence showing that CD36 may serve as a modulator of platelet activation [9]. We have previously provided several lines of evidence for the dependence of sCD40L release on TxA2-dependent platelet activation in T2DM [30]. Our findings in the present study suggest that also sCD36 release is related to TxA2-dependent platelet activation. The contribution of platelets to enhanced CD36 release in T2DM was further strengthened by the observation that patients with T2DM on chronic low-dose aspirin (100 mg/day) treatment, an intervention down-regulating TxA2-dependent platelet activation in this setting [29,31], showed significantly lower plasma CD36 levels than diabetics not on aspirin. Recently, sCD36 has been reported to be a constituent of circulating microparticles derived mainly from platelets in healthy volunteers, rather than being truly soluble as a product of proteolytic cleavage or a distinct isoform [32]. This finding further corroborates the hypothesis of a platelet origin for part of sCD36. We acknowledge that the cross-sectional nature of the study, as well as lack of in vitro confirmation of sCD36 release from any cellular source in culture, does not allow definitive conclusions about platelets as the unique source of plasma CD36. Nevertheless, the in vivo evidence of plasma CD36 levels, being approximately halved in patients on chronic lowdose ASA treatment (which selectively inhibits platelet COX-1dependent thromboxane biosynthesis), and significantly correlated with thromboxane metabolite excretion, strongly supports the hypothesis that sCD36 is largely derived from platelets. As illustrated in Fig. 1, sCD36 is not normalized by ASA treatment, More likely the incomplete down-regulation of sCD36 by aspirin implies that a fraction of plasma CD36 is derived from other cells than platelets, such as monocytes and macrophages. This is confirmed by lack of correlation between sCD36 and urinary thromboxane metabolite excretion, the latter being instead profoundly suppressed by ASA to levels beneath the upper limit of normality of nonaspirinated healthy subjects. We can speculate that use of concurrent drugs other than aspirin, known to modulate thromboxane biosynthesis by acting on the metabolic abnormalities triggering thromboxanedependent platelet activation (drugs affecting glycemic control, statins, antihypertensives), may have contributed to inhibit thromboxane metabolite excretion, while not affecting sCD36 [27]. The presence of macrovascular complications in a large part of the ASA-treated T2DM patients as compared to controls might explain the higher plasma levels of sCD36 in patients with diabetes on aspirin versus healthy subjects. In fact, Handberg and co-workers have recently found that sCD36 is elevated in unstable plaques and that sCD36 correlates with intima media thickness, indicating that CD36 may also be a measure of atherosclerosis burden [24].
1323
In this light, additional antiplatelet strategies in diabetes should be investigated to interrupt CD36-dependent platelet activation. In T2DM, an imbalance between the synthesis of reactive oxygen species (ROS) and the antioxidants has been described. We, first, showed that diabetes is associated with increased formation of F2isoprostanes, bioactive products of arachidonic acid peroxidation, as a correlate of impaired glycemic control and enhanced lipid peroxidation, providing a biochemical link between impaired glycemic control and persistent platelet activation [12]. In the present study, we also observed a highly significant correlation between plasma CD36 levels and the urinary excretion rate of 8-iso-PGF2α, a sensitive marker of in vivo ongoing lipid peroxidation [10], closely related to circulating ox-LDL levels [14]. This is in line with the evidence that a specific CD36-dependent, ox-LDL-mediated signaling pathway is required for platelet activation [5]. Moreover, the relationship between sCD36 and 8-iso-PGF2α is relevant in the light of the fact that 8-isoPGF2α may amplify the aggregation response to subthreshold concentrations of platelet agonists [10]. This might be relevant in a setting in which platelet activation and enhanced free radical formation coincide, such as diabetes mellitus [12]. Furthermore, CD36 appears to be the major receptor responsible for ox-LDL-induced platelet activation that also involves enhanced of release of sCD40L as a marker of platelet-mediated inflammation [33]. Consistent with our findings, oxidized low-density lipoprotein-dependent platelet-derived microvesicles (PMV), the main culprit in the development of thrombosis, have been reported to trigger platelet activation by forming CD36PMV complexes, and to amplify oxidative stress by increasing the level of 8-iso-PGF2α, in a CD36-dependent manner [34]. Moreover, a recent study by Alkhatatbeh et al. suggested that platelet-derived microparticles were the main cellular source of sCD36 in plasma from healthy controls [32]. However, although several lines of evidence suggest that sCD36 in plasma is derived from platelets, the lack of mechanistic data to further support this conclusion has now been pointed out as a limitation of the study. Our findings of a close relationship between plasma levels of CD36 and plasma levels of sCD40L and urinary excretion rate of 8-iso-PGF2α emphasize an important proinflammatory role of platelets, especially under conditions of oxidative stress. Our present findings of elevated sCD36 levels in diabetic patients with long-standing diabetes (>1 year) than in subjects with a diabetes duration b1 year support the hypothesis that sCD36 may have the potential to reflect changes associated with insulin resistance. In fact, insulin resistance, a well-established early predictor of diabetes, is associated with an increased transcription of CD36 in monocytes [35]. Moreover, elevated sCD36 levels are present not only in overt diabetes but also in prediabetic conditions, such as obesity and PCOS [25], and this relationship is in keeping with our positive correlations between sCD36 and fasting plasma glucose and between sCD36 and HbA1c. Furthermore, the evidence that ASA treatment abolishes this significant correlation could lead us to propose that hyperglycemia was responsible for platelet activation, evidencing a key role of CD36 in this biochemical setting. In addition, we assessed plasma CD36 concentrations in the fasting state. A diurnal variation in plasma CD36 levels, according to the glycemic spikes associated with meals, might be hypothesized. However, unpublished data by Handberg and collaborators reported no influence of 3 h euglycemic, hyperinsulinemic clamp on plasma CD36 levels. In conclusion, CD36 may play a key role in settings characterized by enhanced lipid peroxidation and oxidative stress such as diabetes mellitus. In fact, CD36 signaling in response to ox-LDL may induce trapping of macrophages in the arterial intima, promoting atherosclerosis [36]. Moreover, sCD36, as shown by our study, may be linked to platelet activation induced by hyperglycemia, inflammation, and lipid peroxidation. Future studies may elucidate if the incomplete downregulation of sCD36 by aspirin implies that a fraction of plasma
1324
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
CD36 is derived from other cells than platelets such as monocytes and macrophages. Alternatively, additional antiplatelet strategies in diabetes should be investigated to interrupt CD36-dependent platelet activation. Acknowledgments We thank Ellen Lund Sagen for the excellent technical assistance. The contribution to patient recruitment of the General practitioners from the UTAP (Scafa, Italy), including Drs. Laura Creati, Silvio Basile, Rosalba Silvestri, Moreno D'Emilio, and Maria Pia Blasetti, is gratefully acknowledged. We also thank Ms Annalisa Fasciani for her invaluable help with patient blood sampling. The authors declare no conflict of interest. References [1] Ross, R. Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340:115–126; 1999. [2] Collot-Teixeira, S.; Martin, J.; McDermott-Roe, C.; Poston, R.; McGregor, J. L. CD36 and macrophages in atherosclerosis. Cardiovasc. Res. 75:468–477; 2007. [3] Febbraio, M.; Hajjar, D. P.; Silverstein, R. L. CD36: a class B scavenger receptor involved in angiogenesis, atherosclerosis, inflammation, and lipid metabolism. J. Clin. Invest. 108:785–791; 2001. [4] Tandon, N. N.; Lipsky, R. H.; Burgess, W. H.; Jamieson, G. A. Isolation and characterization of platelet glycoprotein IV (CD36). J. Biol. Chem. 264:7570–7575; 1989. [5] Chen, K.; Febbraio, M.; Li, W.; Silverstein, R. L. A specific CD36-dependent signaling pathway is required for platelet activation by oxidized low-density lipoprotein. Circ. Res. 102:1512–1519; 2008. [6] Korporaal, S. J.; Van Eck, M.; Adelmeijer, J.; Ijsseldijk, M.; Out, R.; Lisman, T.; Lenting, P. J.; Van Berkel, T. J.; Akkerman, J. W. Platelet activation by oxidized low density lipoprotein is mediated by CD36 and scavenger receptor-A. Arterioscler. Thromb. Vasc. Biol. 27:2476–2483; 2007. [7] Huang, M. M.; Bolen, J. B.; Barnwell, J. W.; Shattil, S. J.; Brugge, J. S. Membrane glycoprotein IV (CD36) is physically associated with the Fyn, Lyn, and Yes proteintyrosine kinases in human platelets. Proc. Natl. Acad. Sci. U.S.A. 88:7844–7848; 1991. [8] Maschberger, P.; Bauer, M.; Baumann-Siemons, J.; Zangl, K. J.; Negrescu, E. V.; Reininger, A. J.; Siess, W. Mildly oxidized low density lipoprotein rapidly stimulates via activation of the lysophosphatidic acid receptor Src family and Syk tyrosine kinases and Ca2 + influx in human platelets. J. Biol. Chem. 275:19159–19166; 2000. [9] Podrez, E. A.; Byzova, T. V.; Febbraio, M.; Salomon, R. G.; Ma, Y.; Valiyaveettil, M.; Poliakov, E.; Sun, M.; Finton, P. J.; Curtis, B. R.; Chen, J.; Zhang, R.; Silverstein, R. L.; Hazen, S. L. Platelet CD36 links hyperlipidemia, oxidant stress and a prothrombotic phenotype. Nat. Med. 13:1086–1095; 2007. [10] Patrono, C.; Falco, A.; Davì, G. Isoprostane formation and inhibition in atherothrombosis. Curr. Opin. Pharmacol. 5:198–203; 2005. [11] Morrow, J. D.; Hill, K. E.; Burk, R. F.; Nammour, T. M.; Badr, K. F.; Roberts 2nd, L. J. A series of prostaglandin F2-like compounds are produced in vivo in humans by a non-cyclooxygenase, free radical-catalyzed mechanism. Proc. Natl. Acad. Sci. U.S.A. 87:9383–9387; 1990. [12] Davì, G.; Ciabattoni, G.; Consoli, A.; Mezzetti, A.; Falco, A.; Santarone, S.; Pennese, E.; Vitacolonna, E.; Bucciarelli, T.; Costantini, F.; Capani, F.; Patrono, C. In vivo formation of 8-iso-prostaglandin f2alpha and platelet activation in diabetes mellitus: effects of improved metabolic control and vitamin E supplementation. Circulation 99:224–229; 1999. [13] Gopaul, N. K.; Nourooz-Zadeh, J.; Mallet, A. I.; Anggard, E. E. Formation of F2isoprostanes during aortic endothelial cell-mediated oxidation of low density lipoprotein. FEBS Lett. 348:297–300; 1994. [14] Liu, M. L.; Ylitalo, K.; Salonen, R.; Salonen, J. T.; Taskinen, M. R. Circulating oxidized low-density lipoprotein and its association with carotid intima-media thickness in asymptomatic members of familial combined hyperlipidemia families. Arterioscler. Thromb. Vasc. Biol. 24:1492–1497; 2004. [15] Handberg, A.; Levin, K.; Hojlund, K.; Beck-Nielsen, H. Identification of the oxidized low-density lipoprotein scavenger receptor CD36 in plasma: a novel marker of insulin resistance. Circulation 114:1169–1176; 2006.
[16] Sampson, M. J.; Davies, I. R.; Braschi, S.; Ivory, K.; Hughes, D. A. Increased expression of a scavenger receptor (CD36) in monocytes from subjects with type 2 diabetes. Atherosclerosis 167:129–134; 2003. [17] Ferroni, P.; Basili, S.; Falco, A.; Davì, G. Platelet activation in type 2 diabetes mellitus. J. Thromb. Haemost. 2:1282–1291; 2004. [18] Genuth, S.; Alberti, K. G.; Bennett, P.; Buse, J.; Defronzo, R.; Kahn, R.; Kitzmiller, J.; Knowler, W. C.; Lebovitz, H.; Lernmark, A.; Nathan, D.; Palmer, J.; Rizza, R.; Saudek, C.; Shaw, J.; Steffes, M.; Stern, M.; Tuomilehto, J.; Zimmet, P. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 26:3160–3167; 2003. [19] Pignone, M.; Alberts, M. J.; Colwell, J. A.; Cushman, M.; Inzucchi, S. E.; Mukherjee, D.; Rosenson, R. S.; Williams, C. D.; Wilson, P. W.; Kirkman, M. S. Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a scientific statement of the American Heart Association, and an expert consensus document of the American College of Cardiology Foundation. Circulation 121:2694–2701; 2010. [20] Lorenzo, C.; Williams, K.; Hunt, K. J.; Haffner, S. M. The National Cholesterol Education Program—Adult Treatment Panel III, International Diabetes Federation, and World Health Organization definitions of the metabolic syndrome as predictors of incident cardiovascular disease and diabetes. Diabetes Care 30:8–13; 2007. [21] Khuu, H. M.; Robinson, C. A.; Goolsby, K.; Hardy, R. W.; Konrad, R. J. Evaluation of a fully automated high-performance liquid chromatography assay for hemoglobin A1c. Arch. Pathol. Lab. Med. 123:763–767; 1999. [22] Wang, Z.; Ciabattoni, G.; Créminon, C.; Lawson, J.; Fitzgerald, G. A.; Patrono, C.; Maclouf, J. Immunological characterization of urinary 8-epi-prostaglandin F2 alpha excretion in man. J. Pharmacol. Exp. Ther. 275:94–100; 1995. [23] Ciabattoni, G.; Maclouf, J.; Catella, F.; FitzGerald, G. A.; Patrono, C. Radioimmunoassay of 11-dehydrothromboxane B2 in human plasma and urine. Biochim. Biophys. Acta 918:293–297; 1987. [24] Koonen, D. P.; Jensen, M. K.; Handberg, A. Soluble CD36—a marker of the (pathophysiological) role of CD36 in the metabolic syndrome? Arch. Physiol. Biochem. 117:57–63; 2011. [25] Glintborg, D.; Højlund, K.; Andersen, M.; Henriksen, J. E.; Beck-Nielsen, H.; Handberg, A. Soluble CD36 and risk markers of insulin resistance and atherosclerosis are elevated in polycystic ovary syndrome and significantly reduced during pioglitazone treatment. Diabetes Care 31:328–334; 2008. [26] Handberg, A.; Lopez-Bermejo, A.; Bassols, J.; Vendrell, J.; Ricart, W.; FernandezReal, J. M. Circulating soluble CD36 is associated with glucose metabolism and interleukin-6 in glucose-intolerant men. Diab. Vasc. Dis. Res. 6:15–20; 2009. [27] Davì, G.; Catalano, I.; Averna, M.; Notarbartolo, A.; Strano, A.; Ciabattoni, G.; Patrono, C. Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N. Engl. J. Med. 322:1769–1774; 1990. [28] Santilli, F.; Davì, G.; Consoli, A.; Cipollone, F.; Mezzetti, A.; Falco, A.; Taraborelli, T.; Devangelio, E.; Ciabattoni, G.; Basili, S.; Patrono, C. Thromboxane-dependent CD40 ligand release in type 2 diabetes mellitus. J. Am. Coll. Cardiol. 47:391–397; 2006. [29] Inwald, D. P.; McDowall, A.; Peters, M. J.; Callard, R. E.; Klein, N. J. CD40 is constitutively expressed on platelets and provides a novel mechanism for platelet activation. Circ. Res. 92:1041–1048; 2003. [30] Santilli, F.; Rocca, B.; De Cristofaro, R.; Lattanzio, S.; Pietrangelo, L.; Habib, A.; Pettinella, C.; Recchiuti, A.; Ferrante, E.; Ciabattoni, G.; Davì, G.; Patrono, C. Platelet cyclooxygenase inhibition by low-dose aspirin is not reflected consistently by platelet function assays: implications for aspirin "resistance.". J. Am. Coll. Cardiol. 53:667–677; 2009. [31] Davì, G.; Patrono, C. Platelet activation and atherothrombosis. N. Engl. J. Med. 357: 2482–2494; 2007. [32] Alkhatatbeh, M. J.; Mhaidat, N. M.; Enjeti, A. K.; Lincz, L. F.; Thorne, R. F. The putative diabetic plasma marker, soluble CD36, is non-cleaved, non-soluble and entirely associated with microparticles. J. Thromb. Haemost. 9:844–851; 2011. [33] Assinger, A.; Koller, F.; Schmid, W.; Zellner, M.; Koller, E.; Volf, I. Specific binding of hypochlorite-oxidized HDL to platelet CD36 triggers proinflammatory and procoagulant effects. Atherosclerosis 212:153–160; 2010. [34] Wang, H.; Wang, Z. H.; Kong, J.; Yang, M. Y.; Jiang, G. H.; Wang, X. P.; Zhong, M.; Zhang, Y.; Deng, J. T.; Zhang, W. Oxidized low-density lipoprotein-dependent platelet-derived microvesicles trigger procoagulant effects and amplify oxidative stress. Mol. Med.; 2011 Nov 2 (Electronic publication ahead of print). [35] Liang, C. P.; Han, S.; Okamoto, H.; Carnemolla, R.; Tabas, I.; Accili, D.; Tall, A. R. Increased CD36 protein as a response to defective insulin signaling in macrophages. J. Clin. Invest. 113:764–773; 2004. [36] Park, Y. M.; Febbraio, M.; Silverstein, R. L. CD36 modulates migration of mouse and human macrophages in response to oxidized LDL and may contribute to macrophage trapping in the arterial intima. J. Clin. Invest. 119:136–145; 2009.
Contents lists available at SciVerse ScienceDirect
Free Radical Biology & Medicine journal homepage: www.elsevier.com/locate/freeradbiomed
Original Contribution
Plasma levels of soluble CD36, platelet activation, inflammation, and oxidative stress are increased in type 2 diabetic patients Rossella Liani a, Bente Halvorsen b, Simona Sestili a, Aase Handberg c, Francesca Santilli a, Natale Vazzana a, Gloria Formoso a, Pål Aukrust b, Giovanni Davì a,⁎ a b c
Center of Excellence on Aging, University of Chieti “G. d'Annunzio,” Via Colle dell'Ara, 66013, Chieti, Italy Research Institute of Internal Medicine, Oslo University Hospital Rikshospitalet, University of Oslo, Oslo, Norway Department of Clinical Biochemistry, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark
a r t i c l e
i n f o
Article history: Received 10 August 2011 Revised 31 January 2012 Accepted 5 February 2012 Available online 15 February 2012 Keywords: sCD36 Type 2 diabetes Platelet activation Oxidative stress
a b s t r a c t Inflammation, oxidative stress, and platelet activation are involved in type 2 diabetes and its complications. Soluble CD36 (sCD36) has been proposed to early identify diabetics at risk of accelerated atherothrombosis. We aimed at characterizing the platelet contribution to sCD36 in diabetes, by correlating its concentration with the extent of platelet-mediated inflammation and in vivo lipid peroxidation and investigating the effects of low-dose aspirin on these processes. A cross-sectional comparison of sCD36, soluble CD40L (sCD40L) reflecting platelet-mediated inflammation, urinary 11-dehydro-TxB2, and 8-iso-PGF2α, in vivo markers of platelet activation and lipid peroxidation, was performed among 200 diabetic patients (94 of them on aspirin 100 mg/day) and 47 healthy controls. sCD36 levels (median [IQR]: 0.72 [0.31–1.47] vs 0.26 [0.2–0.37], P = 0.003) and urinary 11-dehydro-TxB2 levels (666 [293–1336] vs 279 [160–396], P ≤ 0.0001) were significantly higher in diabetic patients not on aspirin (n = 106) than in healthy subjects. These variables were significantly lower in aspirin-treated diabetics than untreated patients (P b 0.0001). Among patients not on aspirin, those with long-standing diabetes (> 1 year) had significantly higher sCD36 levels in comparison to patients with diabetes duration b 1 year (1.01 [0.62–1.86] vs 0.44 [0.22–1.21], P = 0.001). sCD36 linearly correlated with sCD40L (rho = 0.447; P = 0.0001). On multiple regression analysis, 11-dehydro-TxB2 (β = 0.360; SEM = 0.0001, P = 0.001), 8-iso-PGF2α (β = 0.469; SEM = 0.0001, P b 0.0001), and diabetes duration (β = 0.244; SEM = 0.207, P = 0.017) independently predicted sCD36 levels. sCD36, platelet activation, inflammation, and oxidative stress are increased in type 2 diabetes. Future studies are needed to elucidate if the incomplete down-regulation of sCD36 by low-dose aspirin implies that sCD36 may be derived from tissues other than platelets or if additional antiplatelet strategies in diabetes are necessary to interrupt CD36dependent platelet activation. © 2012 Elsevier Inc. All rights reserved.
Introduction Atherogenesis starts when low-density lipoprotein (LDL) accumulates in the vessel wall at sites of injury and is oxidized to ox-LDL, accumulating into macrophages. The resulting foam cells are characteristic for the early atherosclerotic lesions [1]. The scavenger receptor CD36 is believed to have a critical role in the initiation and progression of atherosclerosis through its ability to bind and internalize modified LDL, facilitating the formation of macrophage foam cells [2].
Abbreviations: 11-dehydro-TxB2, 11-dehydro-thromboxane B2; 8-iso-PG2α, 8-isoprostaglandin F2-alpha; ASA, acetylsalicylic acid; BMI, body mass index; CRP, C-reactive protein; HbA1c, hemoglobin A1c; IL, interleukin; LDL, low-density lipoprotein; ox, oxidized; PCOS, polycystic ovary syndrome; ROS, reactive oxygen species; sCD36, soluble CD36; sCD40L, soluble CD40 ligand; T2DM, type 2 diabetes mellitus. ⁎ Corresponding author. Fax: + 39 0871 541261. E-mail address: [email protected] (G. Davì). 0891-5849/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2012.02.012
CD36 is an 88-kDa glycosylated transmembrane protein expressed in various cell types, including macrophages and endothelial cells [3]. CD36 is involved in physiological and pathological processes in vivo, including atherosclerosis and diabetes [3]. Platelets express scavenger receptors, including CD36 [4], although their functional role in platelet biology has not been clearly defined. A specific CD36-dependent signaling pathway is required for platelet activation by ox-LDL [5]. Ox-LDL is present in the circulation, and platelets can come into contact with ox-LDL and become activated, thereby contributing to the thrombotic process [6]. Platelet CD36 is associated with nonreceptor tyrosine kinases of the Src family [7], implicated in platelet activation by ox-LDL [8]. CD36 engagement may induce an activating signal that is synergistic with signals from other receptors and may result in platelet activation by subthreshold concentrations of physiological agonists. Thus, CD36 joins a growing number of platelet receptors and ligands that may potentiate members of the platelet activation cascade, including Gas6 and its
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
receptors, CD40 ligand, αIIbβ3, Eph kinases and ephrins. Unlike other “coreceptor” ligands, which are primarily localized at the platelet surface during the initial phase of platelet aggregation, CD36 ligands are likely to be presented to the platelet surface before “classical” platelet agonists. Thus, CD36 may serve to “prime” or sensitize the platelet for subsequent activation, and this may contribute to platelet hyperreactivity [9]. Besides ox-LDL, urinary F2-isoprostanes levels represent reliable and sensitive markers of in vivo ongoing lipid peroxidation [10]. Isoprostanes are a family of bioactive compounds produced from arachidonic acid via a free radical-catalyzed mechanism of lipid peroxidation on cell membrane phospholipids or circulating LDL [11]. Moreover, F2-isoprostanes, in particular 8-iso-PGF2α, may amplify the aggregation response to subthreshold concentrations of platelet agonists [10]. This might be relevant to settings in which platelet activation and enhanced free radical formation coincide, such as diabetes mellitus [12]. 8-iso-PGF2α is formed during the in vitro oxidation of LDL [13] and circulating ox-LDL is closely related to isoprostane levels [14]. It has been previously demonstrated that a soluble form of CD36 (sCD36) is present in plasma [15]. In type 2 diabetic patients (T2DM), plasma CD36 is markedly elevated compared to control subjects [15], and it is highly related to insulin resistance and glycemic control. It has been proposed that sCD36 is derived from the surfaces of monocytes and macrophages with increased CD36 expression [15]. Indeed, patients with T2DM have been reported to have enhanced monocyte CD36 expression, potentially reflecting increased posttranscriptional efficiency of CD36 mRNA in response to chronic hyperglycemia [16]. However, platelet activation is a common feature of T2DM [17], and platelets could therefore also contribute to the enhanced levels of sCD36 in these patients. In the present study, we tested the hypothesis that the increased concentrations of sCD36 observed in T2DM may derive, at least in part, from platelet activation. Therefore, the goals of this study were to characterize the platelet contribution to sCD36 in T2DM, by correlating its concentration with the extent of platelet-mediated inflammation and in vivo lipid peroxidation as reflected by enhanced isoprostane formation, and investigating the effects of low-dose aspirin on these processes.
1319
the American Diabetes Association [18,19] and 47 healthy subjects (31 female, 16 male; aged 39 ± 12.8 years) were enrolled in the study. The baseline characteristics of diabetic patients are reported in Table 1. Exclusion criteria were: (1) a recent history (b6 months) of thrombotic events, pregnancy, or lactation; (2) regular use of estroprogestin, iron, antioxidants (vitamins C and E), NSAIDs, antiplatelet agents other than aspirin; (3) clinically significant hepatic, renal, cardiac, or pulmonary insufficiency; (4) history of malignant neoplasms (diagnosed and treated within the last 5 years); (5) autoimmune disorders and type 1 diabetes mellitus. At the time of the study, diabetic patients were being treated by diet alone, insulin alone, diet plus hypoglycemic agent, or hypoglycemic agent plus insulin. Among them, 94 diabetic patients were also being treated with low-dose aspirin (100 mg/day) for primary or secondary cardiovascular prevention [19]. Diabetic patients were examined for the presence of microvascular and macrovascular complications. Of the 200 patients, 27 (13.5%) (20 on ASA treatment) had microvascular complications (nephropathy, detected by the presence of persistent microalbuminuria between 30 and 300 mg/24 h, in at least two of three consecutive 24-h collections, with a eGRF ≥ 90 ml/min/1.73 m 2, as calculated on MDRD formula; retinopathy, based on fundus oculi examination) and 38 (19%) (35 on ASA treatment) had a history or physical examination positive for evidence of macrovascular complications (cerebrovascular disease, coronary heart disease, or peripheral arterial disease). Reasons for lack of aspirin treatment included unwillingness of the patient (one patient) or recent bleeding episodes, eliciting temporary drug withdrawal (2 patients). Diabetic patients with arterial hypertension or hypercholesterolemia were included if well controlled with stable drug therapy: 96 (48%) had arterial hypertension (57 on ASA treatment), defined as current systolic/diastolic blood pressure >130/85 mm Hg; 131 (65.5%) were hypercholesterolemic (71 on ASA treatment), in accordance with the ATPIII criteria [20]. Written informed content was obtained from each subject participating in the study. The local Ethics Committee approved the protocol.
Design of the studies Materials and methods Subjects Two-hundred patients (91 female, 109 male; aged 64.1 ± 8.7 years) with T2DM, as defined in accordance with the criteria of
First, a cross-sectional comparison of circulating sCD36, soluble CD40L (sCD40L) as a marker of platelet-mediated inflammation, urinary 11-dehydro-TxB2, and 8-iso-PGF2α was performed among all patients and controls. All subjects were studied as outpatients after a 12-h fast and performed an overnight urine collection immediately
Table 1 Baseline characteristics of the type 2 diabetic patients. Type 2 diabetes mellitus (n = 200) Variables
Male gender Age (years) BMI (kg/m2) Diabetes duration b 1 year > 1 year Fasting plasma glucose (mmol/L) Hemoglobin A1c (%) Hypertension Hypercholesterolemia Microvascular complications Macrovascular complications
All (n = 200)
Not on ASA (n = 106)
On ASA (n = 94)
n (%) Median Median [IQR]
109 (54.5) 64 [60 to 69.7] 28.1 [25.1 to 31.1]
52 (49) 64 [57.5 to 70] 28.6 [25.4 to 32]
57 (60.6) 66 [61 to 69] 28.1 [24.9 to 30.8]
n (%)
85 (42.5) 115 (57.5) 7.4 [6.6 to 8.5] 6.9 [6.4 to 7.5] 96 (48) 131 (65.5) 27 (13.5) 38 (19)
63 (59.4) 43 (40.6) 7.7 [6.8 to 9.5] 6.9 [6.4 to 7.3] 39 (36.8) 60 (56.6) 7 (6.6) 3 (2.8)
22 (23.4) 72 (76.6) 7.1 [6.2 to 8.3] 6.9 [6.4 to 7.6] 57 (60.6) 71 (75.5) 20 (21.3) 35 (37.2)
Median [IQR] Median [IQR] n (%) n (%) n (%) n (%)
P values 0.118 0.051 0.233 b 0.0001 0.008 0.552 0.001 0.843 0.003 b 0.0001
1320
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
before blood sampling. Urine samples were added with the antioxidant 4-hydroxy-Tempo (1 mM) (Sigma Chemical Co., St. Louis, MO) and stored at -20 °C until extraction. Blood samples were obtained for lipid levels as well as routine blood chemistry to check inclusion and exclusion criteria. To assess the potential influence of long-standing glucose control impairment on sCD36 levels, in the 106 T2DM patients not on ASA treatment, sCD36 levels were also compared between subjects who had a known diabetes duration b1 year, and subjects with longstanding diabetes (>1 year). To test the hypothesis of a platelet origin of sCD36, 94 T2DM patients on aspirin (100 mg/day) were also studied. Finally, we tested the hypothesis of an association among CD36 release in plasma, oxidative stress, and platelet activation in 16 T2DM patients, not on ASA treatment, in whom measurements were repeated after 18–20 weeks. Routine biochemical measurements Fasting plasma glucose was measured by the glucose oxidase method. The HbA1c level was determined by automated highperformance liquid chromatography (HPLC) [21]. Enzyme-linked immunosorbent assay (ELISA) The sCD40L was determined by ELISA (R&D Systems, Minneapolis, MN). The sCD36 was measured using an in-house ELISA [15]. A pool of EDTA plasma, aliquoted and stored at − 80 °C, was applied in seven dilutions in each run and used as a standard concentration curve. Two dilutions of another EDTA pool were used as internal controls, and each control was run in quadruplicate on each ELISA plate. Patient EDTA plasma samples, which were stored at −80 °C, were analyzed in duplicate. Runs were accepted if the controls were within ± 2 SD from mean, and most were within 1 SD. Intraassay CV was 6%, and day-day-assay CV was 16.4%, estimated from internal controls in the study. Standard curves were logtransformed and thereby linear. A few measurements outside the range of the standard curve were calculated from the extrapolated standard curve. In the text, the concentration of sCD36 is given in relative units.
variables significantly related to plasma CD36 in the univariate analysis as covariates. Data are presented as mean (± standard deviation) or as median and interquartile range (IQR) (25th, 75th percentile). Only P values b0.05 were regarded as statistically significant. All tests were twotailed, and analyses were performed using a computer software package (or Statistical Package for the Social Sciences, version 18.0, SPSS Inc., Chicago, IL).
Results Plasma levels of CD36 in T2DM and healthy controls in relation to ASA therapy Soluble CD36 levels were significantly higher in patients with T2DM not on ASA treatment than in healthy subjects (median [IQR]: 0.72 [0.31 to 1.47] vs 0.26 [0.2 to 0.37], P = 0.003), whereas plasma CD36 levels were significantly lower in diabetic patients on ASA treatment when compared with diabetic patients not on ASA (0.39 [0.27 to 0.58] vs 0.72 [0.31 to 1.47], P b 0.0001) but still significantly higher than healthy subjects (0.39 [0.27 to 0.58] vs 0.26 [0.2 to 0.37], P b 0.0001) (Fig. 1A). T2DM patients not on ASA treatment had a significantly higher urinary 11-dehydro-TxB2 excretion rate in comparison with control subjects (666 [293 to 1336] vs 279.5 [160 to 396.5] pg/mg creatinine, P ≤ 0.0001). Urinary 11-dehydro-TxB2 levels were significantly lower in diabetic patients treated with ASA, in comparison with nonaspirin-treated subjects, including diabetic patients not on ASA treatment (146 [100.5 to 247] vs 666 [293 to 1336] pg/mg creatinine, P ≤ 0.0001) and healthy subjects (146 [100.5 to 247] vs 279.5 [160 to 396.5] pg/mg creatinine, P b 0.0001) (Fig. 1B).
Urinary eicosanoid assays Urinary 8-iso-PGF2α and 11-dehydro-TxB2 were measured by previously described radioimmunoassay methods that have been validated by using different antisera and by comparison with gas chromatography/mass spectrometry [22,23]. Statistical analysis Sample size calculation was based on previous data reporting a 4.5-fold difference in plasma CD36 concentration between obese diabetic subjects and lean controls [15]. With 106 patients not on aspirin and 47 controls, the study had a 99% power to detect a difference of at least 3.5-fold between diabetic patients and controls. The Kolmogorov-Smirnov test was used to determine whether each variable had a normal distribution. When necessary, log transformation was used to normalize the data, or appropriate nonparametric tests were used (Mann-Whitney U test; Spearman correlation coefficient). Differences at baseline in categorical variables were analyzed by Fisher's exact test. Comparisons between groups were made with ANOVA with the Scheffè post hoc test to correct for multiple comparisons, with logarithmically transformed data. Comparisons within groups were made with the Wilcoxon test. A multiple linear regression analysis was performed to further quantify the relationship between the sCD36 and the variables in study, with
Fig. 1. Plasma levels of CD36 in T2DM and healthy controls in relation to ASA therapy. Plasma levels of CD36 (A) and urinary 11-dehydro-TxB2 (B) excretion rate in healthy subjects, and in diabetic patients on aspirin and not on aspirin. Data are median (interquartile range). Tx = thromboxane.
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
1321
Plasma levels of CD36 in relation to oxidative stress and markers of platelet activation A significant direct correlation was found between plasma CD36 and urinary 11-dehydro-TxB2 and 8-iso-PGF2α excretion rates (rho = 0.393; P = 0.0003, and rho = 0.481; P = 0.0001, respectively) in diabetic patients not on ASA treatment (Figs. 3A and B). Moreover, a significant linear correlation was found between plasma CD36 and CD40L (rho = 0.447; P = 0.0001) (Fig. 3C). Plasma levels of CD36 in T2DM in relation to metabolic variables, atherosclerotic risk factors, and prostanoid metabolites—Multivariate analyses To further define the relationship between sCD36, metabolic variables, additional atherosclerotic risk factors and urinary prostanoid metabolites in the 106 T2DM not treated with ASA, a multiple regression analysis was performed at baseline in which sCD36 was included
Fig. 2. Plasma levels of CD36 in relation to diabetes duration. Comparison of soluble CD36 in the 106 patients not on ASA treatment, subdivided in two groups: diabetes duration b 1 year and long-standing diabetes (> 1 year). Correlation between fasting plasma glucose and sCD36 (rho = 0.378; p = 0.002) in diabetic patients not on aspirin. Correlation between hemoglobinA1c and sCD36 (rho = 0.255; p = 0.026) in diabetic patients not on aspirin.
Plasma levels of CD36 in relation to diabetes duration T2DM patients not treated with ASA and with long-standing diabetes (>1 year) had significantly higher plasma CD36 levels in comparison with T2DM patients with diabetes duration b1 year (1.01 [0.62 to 1.86] vs 0.44 [0.22 to 1.21], P = 0.001) (Fig. 2A). Consistently, a significant linear correlation was found between plasma CD36 and diabetes duration (rho = 0.347; P = 0.0001). Furthermore in the same group of patients, a significant linear correlation was found between plasma CD36 and fasting blood glucose and HbA1c (rho = 0.378; P = 0.002, and rho = 0.255; P = 0.026, respectively) (Figs. 2B and C). On the contrary, in diabetic patients on ASA, no significant correlation was found between plasma CD36 and HbA1c (rho = 0.050; P = 0.648). No significant relationship was found, either in diabetic patients on ASA or not on ASA, between plasma creatinine and urinary 11dehydro-TxB2 and 8-iso-PGF2α (on ASA subjects rho = -0.127, P = 0.359 and rho = -0.065, P = 0.641, respectively; not on ASA subjects rho = 0.020, P = 0.924 and rho = -0.138, P = 0.520, respectively).
Fig. 3. Plasma levels of CD36 in relation to oxidative stress and markers of plateletactivation. Correlation between urinary 11-dehydro-TxB2 and sCD36 (rho = 0.393; p = 0.0003) in the diabetic patients not on aspirin. Correlation between urinary 8iso-PGF2α and sCD36 (rho = 0.481; p = 0.0001) in diabetic patients not on aspirin. Correlation between plasma levels of CD40L and sCD36 (rho = 0.447; p = 0.0001) in diabetic patients not on aspirin. Tx = thromboxane; PG = prostaglandin.
1322
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
sCD36. In contrast, only 32.3% of T2DM patients with 11-dehydroTxB2 and 8-iso-PGF2α urinary excretion in the first and second quartiles had increased levels of sCD36 (P = 0.004). Plasma levels of CD36 during longitudinal testing The reproducibility of plasma CD36 and CD40L measurements as well as thromboxane biosynthesis and isoprostane formation was assessed by obtaining an additional sample 18-20 weeks later, from 16 patients with T2DM not on ASA treatment. No significant differences in plasma CD36, and CD40L, urinary 11-dehydro-TxB2, and 8iso-PGF2α assessed on the two different occasions (from 0.42 [0.26 to 0.69] to 0.39 [0.22 to 0.56], P = 0.485; from 0.5 [0.16 to 1.06] to 0.42 [0.17 to 1.23] ng/mL, P = 0.570; from 1095 [680.75 to 2052.75] to 1085 [756.51 to 2266.5] pg/mg creatinine, P = 0.245; from 410 [243.75 to 584] to 450.5 [284.75 to 600.64] pg/mg creatinine, P = 0.196, respectively) were detected (Fig. 5A, B, C and D). No detectable change in metabolic control or body weight occurred throughout the 18–20-week period between assessments. Fig. 4. Plasma levels of CD36 in T2DM in relation to metabolic variables, atherosclerotic risk factors and prostanoid metabolites – multivariate analyses. Correlation between 8iso-PGF2α and 11-dehydro-TxB2 excretion rates in type 2 diabetic patients according to quartiles of plasma CD36. Vertical and horizontal lines mark the boundaries of median individual measurements; open circles = first and second quartile of sCD36; solid circles = third and fourth quartile of sCD36. Data are presented on a logarithmic scale. PG = prostaglandin; Tx = thromboxane.
as the dependent variable. Stepwise linear regression yielded a model in which only urinary 11-dehydro-TxB2 (regression coefficient = 0.360; standard error of the mean = 0.0001, P = 0.001) and 8-iso-PGF2α (regression coefficient = 0.469; standard error of the mean = 0.0001, P b 0.0001) and diabetes duration (regression coefficient = 0.244; standard error of the mean = 0.207, P = 0.017) predicted sCD36 levels, independently of gender, fasting blood glucose, and hypercholesterolemia. Thus, as shown in Fig. 4, among 36 diabetic patients with 11-dehydro-TxB2 and 8-iso-PGF2α excretion rates in the third and fourth quartiles, 25 (69.4%) had increased levels of
Discussion Soluble CD36 (sCD36) has been proposed to reflect monocyte/ macrophage and platelet CD36 expression in the metabolic syndrome, with a consistent relationship between insulin resistance and sCD36, potentially being a marker that integrates insulin resistance and atherosclerosis [15]. In fact circulating CD36 may be present as the unbound protein or a peptide fraction thereof or may be present in microparticles shed from cells such as platelets, monocytes/macrophages, or adipocytes after being triggered by various stimuli. Further dedicated studies are required to support this hypothesis [15]. Increased plasma CD36 levels have been described not only in overt T2DM [24] but also in prediabetic conditions such as polycystic ovary syndrome (PCOS) and in subjects with impaired glucose tolerance [25]. sCD36 levels seem to be predicted by chronic low-grade inflammatory state (C-reactive protein [CRP] and interleukin [IL]-6), insulin sensitivity, and BMI [25,26], all involved in the
Fig. 5. Plasma levels of CD36 during longitudinal testing. Urinary 8-iso-PGF2α (A), soluble CD36 (B), urinary 11-dehydro-TxB2 (C), and soluble CD40L (D) in 16 diabetic patients not on aspirin. Individual patient data are also represented in a second sample obtained 18-20 weeks later. Tx = thromboxane; PG = prostaglandin.
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
pathogenesis of diabetes and metabolic syndrome. Thus, sCD36 has been proposed as a simple marker to predict, at an early stage, diabetic subjects at risk of developing accelerated atherothrombosis. Platelets express CD36 [4], and a specific CD36-dependent signaling pathway is required for platelet activation by ox-LDL [5]. Moreover, CD36 may “prime” or sensitize the platelet for subsequent activation, contributing to platelet hyperreactivity [9]. However, the contribution of persistent platelet activation to increased shedding of CD36 has not been previously investigated. T2DM is characterized by persistent thromboxane-dependent platelet activation [17,27]. In the present study we demonstrated that T2DM patients have persistently increased levels of sCD36, related to the impaired metabolic control (plasma glucose levels and HbA1c) and to diabetes duration. Moreover, the highly significant correlation between plasma sCD36 levels and the urinary excretion rate of 11-dehydro-TxB2, a noninvasive index of in vivo platelet activation [27], supports the likelihood of CD36 release during TxA2-dependent platelet activation in T2DM. This relationship is strengthened by the correlation between plasma CD40L and sCD36, considering that activated platelets are the major cellular source of sCD40L in the circulation [28], and that platelet CD40L itself provides a mechanism for platelet activation [29]. Both these positive correlations are in keeping with previously reported evidence showing that CD36 may serve as a modulator of platelet activation [9]. We have previously provided several lines of evidence for the dependence of sCD40L release on TxA2-dependent platelet activation in T2DM [30]. Our findings in the present study suggest that also sCD36 release is related to TxA2-dependent platelet activation. The contribution of platelets to enhanced CD36 release in T2DM was further strengthened by the observation that patients with T2DM on chronic low-dose aspirin (100 mg/day) treatment, an intervention down-regulating TxA2-dependent platelet activation in this setting [29,31], showed significantly lower plasma CD36 levels than diabetics not on aspirin. Recently, sCD36 has been reported to be a constituent of circulating microparticles derived mainly from platelets in healthy volunteers, rather than being truly soluble as a product of proteolytic cleavage or a distinct isoform [32]. This finding further corroborates the hypothesis of a platelet origin for part of sCD36. We acknowledge that the cross-sectional nature of the study, as well as lack of in vitro confirmation of sCD36 release from any cellular source in culture, does not allow definitive conclusions about platelets as the unique source of plasma CD36. Nevertheless, the in vivo evidence of plasma CD36 levels, being approximately halved in patients on chronic lowdose ASA treatment (which selectively inhibits platelet COX-1dependent thromboxane biosynthesis), and significantly correlated with thromboxane metabolite excretion, strongly supports the hypothesis that sCD36 is largely derived from platelets. As illustrated in Fig. 1, sCD36 is not normalized by ASA treatment, More likely the incomplete down-regulation of sCD36 by aspirin implies that a fraction of plasma CD36 is derived from other cells than platelets, such as monocytes and macrophages. This is confirmed by lack of correlation between sCD36 and urinary thromboxane metabolite excretion, the latter being instead profoundly suppressed by ASA to levels beneath the upper limit of normality of nonaspirinated healthy subjects. We can speculate that use of concurrent drugs other than aspirin, known to modulate thromboxane biosynthesis by acting on the metabolic abnormalities triggering thromboxanedependent platelet activation (drugs affecting glycemic control, statins, antihypertensives), may have contributed to inhibit thromboxane metabolite excretion, while not affecting sCD36 [27]. The presence of macrovascular complications in a large part of the ASA-treated T2DM patients as compared to controls might explain the higher plasma levels of sCD36 in patients with diabetes on aspirin versus healthy subjects. In fact, Handberg and co-workers have recently found that sCD36 is elevated in unstable plaques and that sCD36 correlates with intima media thickness, indicating that CD36 may also be a measure of atherosclerosis burden [24].
1323
In this light, additional antiplatelet strategies in diabetes should be investigated to interrupt CD36-dependent platelet activation. In T2DM, an imbalance between the synthesis of reactive oxygen species (ROS) and the antioxidants has been described. We, first, showed that diabetes is associated with increased formation of F2isoprostanes, bioactive products of arachidonic acid peroxidation, as a correlate of impaired glycemic control and enhanced lipid peroxidation, providing a biochemical link between impaired glycemic control and persistent platelet activation [12]. In the present study, we also observed a highly significant correlation between plasma CD36 levels and the urinary excretion rate of 8-iso-PGF2α, a sensitive marker of in vivo ongoing lipid peroxidation [10], closely related to circulating ox-LDL levels [14]. This is in line with the evidence that a specific CD36-dependent, ox-LDL-mediated signaling pathway is required for platelet activation [5]. Moreover, the relationship between sCD36 and 8-iso-PGF2α is relevant in the light of the fact that 8-isoPGF2α may amplify the aggregation response to subthreshold concentrations of platelet agonists [10]. This might be relevant in a setting in which platelet activation and enhanced free radical formation coincide, such as diabetes mellitus [12]. Furthermore, CD36 appears to be the major receptor responsible for ox-LDL-induced platelet activation that also involves enhanced of release of sCD40L as a marker of platelet-mediated inflammation [33]. Consistent with our findings, oxidized low-density lipoprotein-dependent platelet-derived microvesicles (PMV), the main culprit in the development of thrombosis, have been reported to trigger platelet activation by forming CD36PMV complexes, and to amplify oxidative stress by increasing the level of 8-iso-PGF2α, in a CD36-dependent manner [34]. Moreover, a recent study by Alkhatatbeh et al. suggested that platelet-derived microparticles were the main cellular source of sCD36 in plasma from healthy controls [32]. However, although several lines of evidence suggest that sCD36 in plasma is derived from platelets, the lack of mechanistic data to further support this conclusion has now been pointed out as a limitation of the study. Our findings of a close relationship between plasma levels of CD36 and plasma levels of sCD40L and urinary excretion rate of 8-iso-PGF2α emphasize an important proinflammatory role of platelets, especially under conditions of oxidative stress. Our present findings of elevated sCD36 levels in diabetic patients with long-standing diabetes (>1 year) than in subjects with a diabetes duration b1 year support the hypothesis that sCD36 may have the potential to reflect changes associated with insulin resistance. In fact, insulin resistance, a well-established early predictor of diabetes, is associated with an increased transcription of CD36 in monocytes [35]. Moreover, elevated sCD36 levels are present not only in overt diabetes but also in prediabetic conditions, such as obesity and PCOS [25], and this relationship is in keeping with our positive correlations between sCD36 and fasting plasma glucose and between sCD36 and HbA1c. Furthermore, the evidence that ASA treatment abolishes this significant correlation could lead us to propose that hyperglycemia was responsible for platelet activation, evidencing a key role of CD36 in this biochemical setting. In addition, we assessed plasma CD36 concentrations in the fasting state. A diurnal variation in plasma CD36 levels, according to the glycemic spikes associated with meals, might be hypothesized. However, unpublished data by Handberg and collaborators reported no influence of 3 h euglycemic, hyperinsulinemic clamp on plasma CD36 levels. In conclusion, CD36 may play a key role in settings characterized by enhanced lipid peroxidation and oxidative stress such as diabetes mellitus. In fact, CD36 signaling in response to ox-LDL may induce trapping of macrophages in the arterial intima, promoting atherosclerosis [36]. Moreover, sCD36, as shown by our study, may be linked to platelet activation induced by hyperglycemia, inflammation, and lipid peroxidation. Future studies may elucidate if the incomplete downregulation of sCD36 by aspirin implies that a fraction of plasma
1324
R. Liani et al. / Free Radical Biology & Medicine 52 (2012) 1318–1324
CD36 is derived from other cells than platelets such as monocytes and macrophages. Alternatively, additional antiplatelet strategies in diabetes should be investigated to interrupt CD36-dependent platelet activation. Acknowledgments We thank Ellen Lund Sagen for the excellent technical assistance. The contribution to patient recruitment of the General practitioners from the UTAP (Scafa, Italy), including Drs. Laura Creati, Silvio Basile, Rosalba Silvestri, Moreno D'Emilio, and Maria Pia Blasetti, is gratefully acknowledged. We also thank Ms Annalisa Fasciani for her invaluable help with patient blood sampling. The authors declare no conflict of interest. References [1] Ross, R. Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340:115–126; 1999. [2] Collot-Teixeira, S.; Martin, J.; McDermott-Roe, C.; Poston, R.; McGregor, J. L. CD36 and macrophages in atherosclerosis. Cardiovasc. Res. 75:468–477; 2007. [3] Febbraio, M.; Hajjar, D. P.; Silverstein, R. L. CD36: a class B scavenger receptor involved in angiogenesis, atherosclerosis, inflammation, and lipid metabolism. J. Clin. Invest. 108:785–791; 2001. [4] Tandon, N. N.; Lipsky, R. H.; Burgess, W. H.; Jamieson, G. A. Isolation and characterization of platelet glycoprotein IV (CD36). J. Biol. Chem. 264:7570–7575; 1989. [5] Chen, K.; Febbraio, M.; Li, W.; Silverstein, R. L. A specific CD36-dependent signaling pathway is required for platelet activation by oxidized low-density lipoprotein. Circ. Res. 102:1512–1519; 2008. [6] Korporaal, S. J.; Van Eck, M.; Adelmeijer, J.; Ijsseldijk, M.; Out, R.; Lisman, T.; Lenting, P. J.; Van Berkel, T. J.; Akkerman, J. W. Platelet activation by oxidized low density lipoprotein is mediated by CD36 and scavenger receptor-A. Arterioscler. Thromb. Vasc. Biol. 27:2476–2483; 2007. [7] Huang, M. M.; Bolen, J. B.; Barnwell, J. W.; Shattil, S. J.; Brugge, J. S. Membrane glycoprotein IV (CD36) is physically associated with the Fyn, Lyn, and Yes proteintyrosine kinases in human platelets. Proc. Natl. Acad. Sci. U.S.A. 88:7844–7848; 1991. [8] Maschberger, P.; Bauer, M.; Baumann-Siemons, J.; Zangl, K. J.; Negrescu, E. V.; Reininger, A. J.; Siess, W. Mildly oxidized low density lipoprotein rapidly stimulates via activation of the lysophosphatidic acid receptor Src family and Syk tyrosine kinases and Ca2 + influx in human platelets. J. Biol. Chem. 275:19159–19166; 2000. [9] Podrez, E. A.; Byzova, T. V.; Febbraio, M.; Salomon, R. G.; Ma, Y.; Valiyaveettil, M.; Poliakov, E.; Sun, M.; Finton, P. J.; Curtis, B. R.; Chen, J.; Zhang, R.; Silverstein, R. L.; Hazen, S. L. Platelet CD36 links hyperlipidemia, oxidant stress and a prothrombotic phenotype. Nat. Med. 13:1086–1095; 2007. [10] Patrono, C.; Falco, A.; Davì, G. Isoprostane formation and inhibition in atherothrombosis. Curr. Opin. Pharmacol. 5:198–203; 2005. [11] Morrow, J. D.; Hill, K. E.; Burk, R. F.; Nammour, T. M.; Badr, K. F.; Roberts 2nd, L. J. A series of prostaglandin F2-like compounds are produced in vivo in humans by a non-cyclooxygenase, free radical-catalyzed mechanism. Proc. Natl. Acad. Sci. U.S.A. 87:9383–9387; 1990. [12] Davì, G.; Ciabattoni, G.; Consoli, A.; Mezzetti, A.; Falco, A.; Santarone, S.; Pennese, E.; Vitacolonna, E.; Bucciarelli, T.; Costantini, F.; Capani, F.; Patrono, C. In vivo formation of 8-iso-prostaglandin f2alpha and platelet activation in diabetes mellitus: effects of improved metabolic control and vitamin E supplementation. Circulation 99:224–229; 1999. [13] Gopaul, N. K.; Nourooz-Zadeh, J.; Mallet, A. I.; Anggard, E. E. Formation of F2isoprostanes during aortic endothelial cell-mediated oxidation of low density lipoprotein. FEBS Lett. 348:297–300; 1994. [14] Liu, M. L.; Ylitalo, K.; Salonen, R.; Salonen, J. T.; Taskinen, M. R. Circulating oxidized low-density lipoprotein and its association with carotid intima-media thickness in asymptomatic members of familial combined hyperlipidemia families. Arterioscler. Thromb. Vasc. Biol. 24:1492–1497; 2004. [15] Handberg, A.; Levin, K.; Hojlund, K.; Beck-Nielsen, H. Identification of the oxidized low-density lipoprotein scavenger receptor CD36 in plasma: a novel marker of insulin resistance. Circulation 114:1169–1176; 2006.
[16] Sampson, M. J.; Davies, I. R.; Braschi, S.; Ivory, K.; Hughes, D. A. Increased expression of a scavenger receptor (CD36) in monocytes from subjects with type 2 diabetes. Atherosclerosis 167:129–134; 2003. [17] Ferroni, P.; Basili, S.; Falco, A.; Davì, G. Platelet activation in type 2 diabetes mellitus. J. Thromb. Haemost. 2:1282–1291; 2004. [18] Genuth, S.; Alberti, K. G.; Bennett, P.; Buse, J.; Defronzo, R.; Kahn, R.; Kitzmiller, J.; Knowler, W. C.; Lebovitz, H.; Lernmark, A.; Nathan, D.; Palmer, J.; Rizza, R.; Saudek, C.; Shaw, J.; Steffes, M.; Stern, M.; Tuomilehto, J.; Zimmet, P. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 26:3160–3167; 2003. [19] Pignone, M.; Alberts, M. J.; Colwell, J. A.; Cushman, M.; Inzucchi, S. E.; Mukherjee, D.; Rosenson, R. S.; Williams, C. D.; Wilson, P. W.; Kirkman, M. S. Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a scientific statement of the American Heart Association, and an expert consensus document of the American College of Cardiology Foundation. Circulation 121:2694–2701; 2010. [20] Lorenzo, C.; Williams, K.; Hunt, K. J.; Haffner, S. M. The National Cholesterol Education Program—Adult Treatment Panel III, International Diabetes Federation, and World Health Organization definitions of the metabolic syndrome as predictors of incident cardiovascular disease and diabetes. Diabetes Care 30:8–13; 2007. [21] Khuu, H. M.; Robinson, C. A.; Goolsby, K.; Hardy, R. W.; Konrad, R. J. Evaluation of a fully automated high-performance liquid chromatography assay for hemoglobin A1c. Arch. Pathol. Lab. Med. 123:763–767; 1999. [22] Wang, Z.; Ciabattoni, G.; Créminon, C.; Lawson, J.; Fitzgerald, G. A.; Patrono, C.; Maclouf, J. Immunological characterization of urinary 8-epi-prostaglandin F2 alpha excretion in man. J. Pharmacol. Exp. Ther. 275:94–100; 1995. [23] Ciabattoni, G.; Maclouf, J.; Catella, F.; FitzGerald, G. A.; Patrono, C. Radioimmunoassay of 11-dehydrothromboxane B2 in human plasma and urine. Biochim. Biophys. Acta 918:293–297; 1987. [24] Koonen, D. P.; Jensen, M. K.; Handberg, A. Soluble CD36—a marker of the (pathophysiological) role of CD36 in the metabolic syndrome? Arch. Physiol. Biochem. 117:57–63; 2011. [25] Glintborg, D.; Højlund, K.; Andersen, M.; Henriksen, J. E.; Beck-Nielsen, H.; Handberg, A. Soluble CD36 and risk markers of insulin resistance and atherosclerosis are elevated in polycystic ovary syndrome and significantly reduced during pioglitazone treatment. Diabetes Care 31:328–334; 2008. [26] Handberg, A.; Lopez-Bermejo, A.; Bassols, J.; Vendrell, J.; Ricart, W.; FernandezReal, J. M. Circulating soluble CD36 is associated with glucose metabolism and interleukin-6 in glucose-intolerant men. Diab. Vasc. Dis. Res. 6:15–20; 2009. [27] Davì, G.; Catalano, I.; Averna, M.; Notarbartolo, A.; Strano, A.; Ciabattoni, G.; Patrono, C. Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N. Engl. J. Med. 322:1769–1774; 1990. [28] Santilli, F.; Davì, G.; Consoli, A.; Cipollone, F.; Mezzetti, A.; Falco, A.; Taraborelli, T.; Devangelio, E.; Ciabattoni, G.; Basili, S.; Patrono, C. Thromboxane-dependent CD40 ligand release in type 2 diabetes mellitus. J. Am. Coll. Cardiol. 47:391–397; 2006. [29] Inwald, D. P.; McDowall, A.; Peters, M. J.; Callard, R. E.; Klein, N. J. CD40 is constitutively expressed on platelets and provides a novel mechanism for platelet activation. Circ. Res. 92:1041–1048; 2003. [30] Santilli, F.; Rocca, B.; De Cristofaro, R.; Lattanzio, S.; Pietrangelo, L.; Habib, A.; Pettinella, C.; Recchiuti, A.; Ferrante, E.; Ciabattoni, G.; Davì, G.; Patrono, C. Platelet cyclooxygenase inhibition by low-dose aspirin is not reflected consistently by platelet function assays: implications for aspirin "resistance.". J. Am. Coll. Cardiol. 53:667–677; 2009. [31] Davì, G.; Patrono, C. Platelet activation and atherothrombosis. N. Engl. J. Med. 357: 2482–2494; 2007. [32] Alkhatatbeh, M. J.; Mhaidat, N. M.; Enjeti, A. K.; Lincz, L. F.; Thorne, R. F. The putative diabetic plasma marker, soluble CD36, is non-cleaved, non-soluble and entirely associated with microparticles. J. Thromb. Haemost. 9:844–851; 2011. [33] Assinger, A.; Koller, F.; Schmid, W.; Zellner, M.; Koller, E.; Volf, I. Specific binding of hypochlorite-oxidized HDL to platelet CD36 triggers proinflammatory and procoagulant effects. Atherosclerosis 212:153–160; 2010. [34] Wang, H.; Wang, Z. H.; Kong, J.; Yang, M. Y.; Jiang, G. H.; Wang, X. P.; Zhong, M.; Zhang, Y.; Deng, J. T.; Zhang, W. Oxidized low-density lipoprotein-dependent platelet-derived microvesicles trigger procoagulant effects and amplify oxidative stress. Mol. Med.; 2011 Nov 2 (Electronic publication ahead of print). [35] Liang, C. P.; Han, S.; Okamoto, H.; Carnemolla, R.; Tabas, I.; Accili, D.; Tall, A. R. Increased CD36 protein as a response to defective insulin signaling in macrophages. J. Clin. Invest. 113:764–773; 2004. [36] Park, Y. M.; Febbraio, M.; Silverstein, R. L. CD36 modulates migration of mouse and human macrophages in response to oxidized LDL and may contribute to macrophage trapping in the arterial intima. J. Clin. Invest. 119:136–145; 2009.