Tohoku J. Exp. Med., 2007, 213, 129-137 Efflux and PLTP Activity in Diabetes with CAD 129

Increased Phospholipid Transfer Activity Associated with the Impaired Cellular Cholesterol Efflux in Type 2 Diabetic Subjects with

1,2 2 2 2 3 NEBIL ATTIA, AMEL NAKBI, MAHA SMAOUI, RAJA CHAABA, PHILIPPE MOULIN, 4 4 5 SONIA HAMMAMI, KHALDOUN BEN HAMDA, FRANÇOISE CHANUSSOT and 2 MOHAMED HAMMAMI

1Biology Department, Faculty of Sciences, University November 7th at Carthage, Bizerte, Tunisia 2Biochemistry Laboratory, UR 08-39, Faculty of Medicine, University of Monastir, Monastir, Tunisia 3U11 Cardiovascular Unit, Louis Pradel Hospital and INSERM U585, Lyon, France 4Departments of Internal Medicine (SH) and Cardiology (KBH), Monastir Hospital, Monastir, Tunisia 5INSERM U476, Faculty of Medecine la Timone, University Aix-Marseille II, Marseille, France

ATTIA, N., NAKBI, A., SMAOUI, M., CHAABA, R., MOULIN, P., HAMMAMI, S., HAMDA, K.B., CHANUSSOT, F. and HAMMAMI, M. Increased Phospholipid Transfer Protein Activity Associated with the Impaired Cellular Cholesterol Efflux in Type 2 Diabetic Subjects with Coronary Artery Disease. Tohoku J. Exp. Med., 2007, 213 (2), 129-137 ── Reverse cholesterol transport (RCT) is the pathway, by which the excess of cholesterol is removed from peripheral cells to the liver. An early step of RCT is the efflux of free cholesterol from cell membranes that is mediated by high-density (HDL). Phospholipid transfer protein (PLTP) transfers phospholipids between -B-containing lipo- (i.e., and very low-density lipoproteins) and HDL. PLTP contrib- utes to the HDL maturation and increases the ability of HDL to extract the cellular choles- terol. It is known that RCT is impaired in type 2 diabetic patients, especially when cardiovascular complication is associated with. In this study, we measured the serum capacity that promotes cellular cholesterol efflux and the plasma PLTP activity in type 2 diabetic patients with coronary artery disease (CAD) (n = 35), those without CAD (n = 24), and 35 healthy subjects as a sex- and age-matched control. In patients with CAD, plasma level was higher compared to controls (p < 0.01) and HDL-cholesterol was lower (p < 0.01 vs control and the patients without CAD). In diabetic patients with or without CAD, PLTP activity was consistently increased, compared to controls, while cellu- lar cholesterol efflux activity was decreased by 20% (p < 0.001) or 13.5% (p < 0.01), respectively. In conclusion, plasma PLTP activity was increased in type 2 diabetic patients with or without CAD, which could impair cellular cholesterol removal and might acceler- ate in diabetic patients. ──── cholesterol efflux; PLTP activity; Type 2 diabetes; coronary artery disease © 2007 Tohoku University Medical Press

Received May 2, 2007; revision accepted for publication August 20, 2007. Correspondence: Dr. Nebil Attia, Ph.D., Department of Biology, Sciences Faculty of Bizerte, 7021 Jarzouna, Tunisia. e-mail: [email protected] 129 130 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 131

Type 2 diabetes (non-insulin dependent) is Few studies have described the serum capac- associated with cardiovascular risk including high ity to promote cellular cholesterol in type 2 diabe- plasma (TG) and low high-density tes with coronary artery disease (CAD). It was cholesterol (HDL-C) (West et al. reported that LpAI: AII particles are remarkably 1983). The negative relationship between the involved at this step of RCT, especially when type HDL-C concentration and the risk of cardiovascu- 2 and CAD coexist (Taskinen 1990; Cavallero et lar disease (Gordon and Rifkind 1989) is thought al. 1995; Syvänne et al. 1996). to be attributable, at least in part, to the key role All these data show that the PLTP is associ- of HDL in reverse cholesterol transport (RCT) ated with the atherosclerosis, but do not prove from peripheral tissues to the liver for excretion that PLTP is responsible for the impairment of the (Fielding and Fielding 1995; Tall and Wang cellular cholesterol efflux in diabetes. In this 2000). The first step of RCT is efflux of free cho- work we attempted to clarify the role of the PLTP lesterol from cell membranes to the extracellular in the extraction of cholesterol from Fu5AH rat medium. Numbers of studies have established hepatoma cells in type 2 diabetes with or without that at least three mechanisms are involved: a CAD. non-specific and relatively inefficient aqueous diffusion pathway which operates in all cells, and PATIENTS AND METHODS two regulated pathways, which are modulated by Patients phospholipid (PL)-containing acceptor particles A total of 59 Tunisian patients with diabetes melli- (such as HDL) or -poor (apo) tus (35 with CAD and 24 without; respectively D+C+ (Yancey et al. 2003). On the other hand, it has and D+C- groups) were recruited in the Department of been shown that plasma preβ-HDL is elevated in Cardiology and Internal Medicine in the University hypertriglyceridemia (Ishida et al. 1987). These Hospital of Monastir, and 35 healthy subjects served as a particles are considered to be the initial acceptors control group (age range: 40-74 years). Written or oral of cellular cholesterol (Castro and Fielding 1988), consent was obtained from all the patients and healthy and cellular cholesterol efflux, measured using subjects before the study. Written consents were not pos- Fu5AH cells, is positively correlated with sible in all cases because the majority of eligible subjects preβ-HDL (Dullart and Van Tol 2001). One of in our study were unable to read or write. The protocol processes of cellular cholesterol efflux is mediated was approved by the Ethics Committee of Monastir by the ATP-binding cassette transporter A1 Hospital (CELHM). (ABCA1). The diagnosis of diabetes was based on a previous The early step of RCT is also controlled by history of diabetes on American Diabetes Association phospholipid transfer protein (PLTP) which trans- criteria (The Expert Committee on the Diagnosis and fers PLs between apoB-containing lipoproteins Classification of Diabetes Mellitus 1997). The diabetic and HDL, and catalyzing HDL conversion reac- patients without CAD had a normal ECG and no history tions. PLTP may therefore participate in RCT and or clinical signs of CAD with a maximal negative exer- cise test. The patients with CAD were defined as those cholesterol efflux by regenerating nascent discoi- having clinical history of stable angina pectoris, previous dal HDL (ndHDL) during HDL conversion, and acute coronary syndromes with or without ST segment facilitating the lipidation of ndHDL in the suben- elevation, and this CAD was confirmed by coronary dothelial space (Ohashi et al. 2005). At the same angiography. The extent of CAD was assessed by the time, PLTP interacts with and stabilizes ABCA1, number of coronary vessels with more than 50% reduc- and enhances cholesterol efflux from cells (Oram tion in the luminal diameter. Exclusion criteria were et al. 2003). Generally, PLTP is now considered taking insulin or lipid lowering drugs (apart from a as an atherosclerosis risk factor in animals taking by most of CAD patients), having a renal or liver (Vikstedt et al. 2007) and human with obesity, failure or thyroid disease, alcohol consumption 3 days diabetes and/or coronary artery disease (Jiang et prior to the study or less, a body mass index (BMI) more al. 2002; Cheung et al. 2006; de Vries et al. 2006). than 35 kg/m² and glycated hemoglobin (Hb A1c) more 130 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 131

than 12%. Plasma triglyceride concentration was under ments, cells were seeded in six-well plates at a density of 10 mmol/l in type 2 diabetes and under 2.3 mmol/l in 45 000-50 000 cells per well and grown in MEM wit 5% healthy controls. Post menopausal women had no hor- calf serum for two days. Radiolabeled cholesterol mone replacement therapy. Hypertension in diabetic ([1,2-3H]cholesterol, TRK 330, Amersham, France) was population was diagnosed according to predetermined added to the cells: a tracer amount was first mixed into blood pressure level (systolic blood pressure > 130 25% calf serum in MEM, then diluted with MEM to 5% mmHg and/or diastolic blood pressure level > 85 mmHg) final serum concentration. The cells were grown in the (The Sixth Report of the Joint National Committee on presence of radiolabeled for two more days to obtain prevention, detection, evaluation, and treatment of high confluent monolayers, and to allow for radiolabeled cho- blood pressure 1997) or the patients were on antihyper- lesterol to exchange into all cellular pool. The labeling tensive therapy. Body mass index (BMI) was calculated medium was finally replaced by MEM containing 0.5% using the formula: weight (kg)/height² (m²). Obesity was bovine serum albumin for 18 to 20 hrs before measuring defined as BMI > 30 kg/m². The waist-to-hip ratio was of cholesterol efflux to further ensure that pools were calculated from measurements of the waist circumfer- equilibrated. ence taken at the midpoint between the umbilicus and The capacity of serum to promote efflux from the xiphoid and the hip circumference, at the widest point labeled cells was assayed by adding individual samples around the hips, respectively. Blood samples were drawn of the patients’ serum to a final concentration of 5% and after subjects had fasted overnight (12 hrs) into tubes quantifying the amount of radiolabeled cholesterol containing EDTA and plasma was immediately separated release over 4 hrs. At the end of the efflux period, the by centrifugation. Samples used for PLTP activity or medium was collected into ice-chilled tubes and centri- cholesterol efflux were immediately stored at −80°C for fuged in the cold for 5 min at 2,000 rpm to remove cells. at most 3 months. In patients with CAD, the blood sam- Labeled cholesterol release was measured in an aliquot ples were not collected during acute coronary syndrome. of the medium by standard liquid scintillation counting. The main relevant clinical characteristics of patients and Cell monolayers were washed with phosphate-buffered controls are given in Table 1. saline (PBS), were extracted overnight with iso- propanol at room temperature and radioactivity was Laboratory procedures quantified in an aliquot of the extract. Fractional efflux Plasma glucose concentrations were determined by was calculated by dividing radioactivity released into using an enzymatic kit (glucose oxidase). Hb A1c values medium by the total label in each well (medium + cells), were determined using an ion-exchange microcolumn and was expressed as arbitrary units corresponding to chromatographic procedure (Biosystems, Barcelona, efflux as a percent of total. The reproducibility of the Spain) where normal values are 4.2-6.2%. Total choles- response obtained with different batches of cells and terol and TG concentrations were measured on RA 1000 labeling media, was verified by inclusion of a standard analyser by enzymatic methods using Randox reagents pool of human serum in each efflux experiment. All and HDL phospholipids using enzymatic commercial kit assays were effectuated in triplicate. Coefficients of (Biomerieux). HDL cholesterol and phospholipids were variation were 6.5% inter-assay, and 5.0% intra-assay. quantified after precipitation of apoB containing lipopro- teins with magnesium/phosphotungstate. LDL cholester- PLTP activity assay ol was calculated by the Friedwald formula (Friedwald et The plasma phospholipid transfer activity mediated al. 1972). Apolipoproteins A-I and B were measured by by PLTP was determined by measuring the transfer of rate nephelometry on the Beckman reagents. [14C]-phosphatidylcholine (PC) from phospholipid lipo- somes (PL donors) to HDL (Acceptors) (Albers et al. Cell culture and cholesterol efflux assay 1995). The phospholipid liposomes were prepared as Cell cholesterol efflux was measured in the presence described by Damen, Regts and Scherfof (Damen et al. of a 5% dilution of serum, essentially as described by De 1982) with some modifications: the trace-labeled lipo- la Llera Moya et al. (1994). Fu5AH cells were grown in somes were prepared by the sonication of Soya PC (125 Eagle’s minimal essential medium (MEM) supplemented nmol PL/assay) with of [14C]-dipalmitoyl PC (6,000 with 5% bovine calf serum (Boehringer, Meylan, France) dpm/assay, New England Nuclear, NEC 682, NEC, and added with penicillin-streptomycin. For experi- Tokyo) and 1.5 ml Tris buffer (10 mM Tris, 150 mM 132 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 133

NaCl, 1 mM EDTA, 10 nM BHT) pH 7.4. Each tube in Statistical analysis the PLTP activity assay contains 50 μl plasma (source of Values are expressed as mean ± S.D. Triglyceride PLTP) that had been diluted 1:50 by Tris buffer and concentrations were normalized by logarithmic transfor- added 50 μl radiolabeled-liposomes, in presence of 50 μl mation when included in statistical calculations. HDL as acceptor (157 nmol phospholipid) and Tris Differences between groups were compared by one-way buffer up to a final volume of 400μ l. Assay tubes were analysis of variance (ANOVA) with post hoc test incubated at 37°C for 15 min in a shaking water bath, (Bonferonni/Dunn test). Bivariate correlation coeffi- and the reaction stopped by transferring the tubes to an cients were evaluated by linear regression analysis. A ice bath. The donor and acceptor particles were immedi- two-sided p value < 0.05 was considered significant. All ately separated by the addition of 200 μl of carrier plas- statistics were performed using the Statview® package ma protein (heat-inactivated at 56°C for 30 min and (version 5, SAS institute Inc.). diluted 2:5 v/v in 10 mM Tris buffer), 100 μl of 2% dex- tran sulphate combined with 100 μl of 1 M MgCl2. After RESULTS that, liposomes were sedimented by centrifugation for 15 Clinical characteristics and lipid profile min at 3,000 rpm. The radioactivity transferred to HDL Age was significantly higher in D+C+ group was counted in 200 μl aliquot of the supernatant. The compared to controls (respectively 56.75 ± 5.83 assay was not influenced by the phospholipids transfer vs 52.74 ± 7.81 yrs; p < 0.05). Waist-to-hip ratio promoting properties of CETP, as CETP did not transfer was also increased in D+C- and D+C+ patients phospholipids from the liposomes to the HDL. Plasma (p < 0.05 and p < 0.001 vs controls; respectively). PLTP activity was expressed in arbitrary unit (AU) and D+C- group but not D+C+ was mildly hyperten- calculated as the percent of radiolabeled phosphatidyl- choline transferred from liposomes to HDL in 15 min. sive (Table 1). Each sample was assayed in triplicate. Each series also Compared to controls, triglyceridemia was contained triplicate blank tubes without plasma (sponta- increased significantly only in D+C+ group (p < neous transfer without PLTP). The interassay coefficient 0.01) (Table 2). Total as well as LDL-cholesterol of variation was 5.6%. were not significantly different in all groups. Apo B was increased, but not significantly, in diabetic

TABLE 1. Clinical characteristics of subjects. Controls Diabetics Diabetics with CAD Number of Subjects 35 24 35 Sex (M/F) 19/16 12/12 27/8 Age (yrs) 52.74 ± 7.81 54.40 ± 5.54 56.75 ± 5.83* BMI (kg/m²) 26.66 ± 4.30 29.04 ± 4.53 28.24 ± 3.69 WHR 0.91 ± 0.09 0.96 ± 0.04* 0.97 ± 0.058** Duration of diab (yrs) - 8.19 ± 4.64 8.27 ± 5.74 Glycemia (mmol/l) 5.26 ± 0.47 10.84 ± 3.81*** 10.97 ± 3.43*** Hb A1c (%) 5.59 ± 2.16 8.76 ± 1.97** 9.02 ± 1.81*** SBP (mmHg) 125.7 ± 15.8 141.2 ± 27.4** 128.1 ± 16.0*# DBP (mmHg) 76.5 ± 0.82 82.4 ± 10.3* 76.7 ± 11.1

Data are mean ± S.D. (or numbers). Statistical analyses were made by one-way analysis of variance (ANOVA) with post hoc test (Bonferonni/Dunn test). *p < 0.05; **p < 0.01; ***p < 0.001, differences from controls and #p < 0.05, difference from diabetic group without coronary artery disease. CAD, coronary artery disease; BMI, body mass index; WHR, waist to hip ratio; Hb A1c, glycated hemoglobin A1; SBP, systolic blood pressure; DSP, diastolic blood pressure. CAD, coronary artery disease; BMI, body mass index; WHR, waist to hip ratio; Hb A1c, glycated hemoglobin A1; SBP, systolic blood pressure; DSP, diastolic blood pressure. 132 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 133

TABLE 2. Lipidemic profiles and HDL composition of subjects. Controls Diabetics Diabetics with CAD Number of Subjects 35 24 35 TG (mmol/l) 1.43 ± 0.75 1.96 ± 1.35 2.16 ± 1.11** Chol (mmol/l) 4.92 ± 1.26 5.16 ± 1.15 5.18 ± 1.25 LDL-C (mmol/l) 3.31 ± 1.24 3.32 ± 1.32 3.43 ± 0.80 Apo B (g/l) 1.07 ± 0.47 1.06 ± 0.32 1.12 ± 0.38 Apo A-I (g/l) 1.39 ± 0.27 1.31 ± 0.22 1.28 ± 0.24 Apo B/A-I ratio 1.07 ± 0.47 1.06 ± 0.32 1.12 ± 0.38 HDL-C (mmol/l) 0.96 ± 0.23 0.88 ± 0.20 0.73 ± 1.13***## HDL-PL (mmol/l) 0.90 ± 0.27 0.82 ± 0.26 0.81 ± 0.20

Data are mean ± S.D. Statistical analyses were performed by one-way analysis of variance (ANOVA) with post hoc test (Bonferonni/Dunn test). *p < 0.05; **p < 0.01; ***p < 0.01 differences from controls and ##p < 0.01, difference from diabetic group without coronary artery disease. CAD, coronary artery disease; TG, plasma triglycerides; Chol., plasma cholesterol, LDL-C; low density lipoprotein cholesterol; Apo B, ; Apo A-I, apolipoprotein A-I; HDL-C, high density lipoprotein cholesterol; HDL-PL, high density lipoprotein phospholipids.

patients with CAD. Alternatively, apo A-I which efflux was compared between the three groups. was decreased in the same group (NS) in con- Compared to controls diabetic patients without comitance with HDL-C: 0.92 ± 0.24; 0.92 ± 0.25 CAD had a decreased cholesterol efflux by 13.5% and 0.76 ± 0.18 mmol/l; respectively in controls, and those with CAD by 20%: controls, 30.11 ± D+C- and D+C+ groups (p < 0.01; D+C+ group 5.90; D+C-, 26.44 ± 4.42 and D+C+, 24.77 ± vs controls). However, the HDL-phospholipids 4.30% (p < 0.01 and p < 0.001 controls vs D+C- did not significantly differ between the three and D+C+ groups; respectively) (Fig. 1). On the groups (Table 2). opposite, PLTP activity was significantly in- creased both in D+C- and D+C+ groups: controls, Correlations between lipid and clinical param- 18.13 ± 6.87; D+C-, 21.67 ± 5.39 and D+C+, eters 22.09 ± 5.71 AU (p < 0.05 and p < 0.01, respec- In total population, waist to hip ratio was tively). positively correlated to age (r = 0.446; p < 0.0001), In total population, the capacity of serum to to glycemia (r = 0.276; p = 0.0107) and to logTG promote cholesterol efflux was closely related to (r = 0.307; p = 0.0040); and inversely correlated HDL-phospholipids (r = +0.403; p = 0.0004) as to HDL-C (r = −0.223; p = 0.041). On the other well as to apo A-I concentrations (r = +0.282; p = hand, BMI was related to LDL-C (r = 0.232; 0.017). p = 0.0295) and Hb A1c to log TG (r = 0.468; In the whole studied population, PLTP activ- p = 0.0003). As expected, log TG was positively ity was closely associated with cholesterol efflux correlated to apo B (r = 0.271; p = 0.019) and level (r = +0.391; p = 0.0003). Analysis of corre- inversely associated to apo A-I (r = −0.212; lation in each group has shown that such correla- p = 0.048) as well as to HDL-C (r = −0.355; tion was maintained in controls (r = +0.827; p < p = 0.0003). 0.0001) but disappeared in D+C- and D+C+ groups (r = +0.154 and r = 0.134 respectively; Whole serum promoted cholesterol efflux, lipid NS) (Fig. 2). profile and PLTP activity Serum ability to promote cellular cholesterol 134 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 135

Fig. 1. Distributions of Cholesterol efflux and PLTP activity. CAD, coronary artery disease. Data were analyzed using one-way analysis of variance (ANOVA) with post hoc test (Bonferonni/ Dunn test). *p < 0.05, **p < 0.01, ***p < 0.001, difference from controls.

Fig. 2. Correlations between cellular cholesterol efflux and PLTP activity. CAD, coronary artery disease.

were higher in all diabetic patients (Howard 1987; DISCUSSION Attia et al. 1995, 1997; Chaaba et al. 2005a). We previously showed that in diabetic Nevertheless, the increase of TG was more patients with no signs of macrovascular damage, noticeable in D+C+ group compared to healthy the capacity of serum to promote cholesterol from subjects. It was reported by Fournier et al. (2001) Fu5AH rat hepatoma cells was significantly that the capacity of serum from hypertriglyceride- decreased, which was restored to control level by mic subjects to promote cholesterol efflux out of lipid lowering bezafibrate treatment (Autran et al. Fu5AH model cells and macrophages (Fournier et 2000). In the present study, we have confirmed al. 2003) is not decreased, despite lower HDL this finding, in agreement with other researchers cholesterol. These studies did not examine the (Cavallero et al. 1995; Syvänne et al. 1996; Brites relationship between cellular cholesterol extrac- et al. 1999), but not with Dullaart and van Tol tion in diabetes and coronary alteration. Our (2001a). results dealing with cholesterol efflux might be In agreement with earlier studies (Cavallero attributed to diabetes and its cardiovascular com- et al. 1995; Syvänne et al. 1996), we found that plication. the cellular cholesterol efflux was affected by dia- Data show that plasma preβ-HDL, initial and betes, particularly when it was associated with efficient acceptors of cellular cholesterol are CAD. Both HDL-PL and apo A-I stimulated elevated in hypertriglyceridemia (von Eckardstein cellular cholesterol extraction. et al. 1996). Since our patients are hypertriglyc- According to previous studies in fasting as eridemic, we expected a similar effect on cellular well as in postprandial states, plasma triglycerides cholesterol efflux capacity by generating 134 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 135

preβ-HDL particles. But decreased cholesterol CETP with cellular cholesterol efflux level nor efflux level in D+C- and D+C+ patients is likely with PLTP activity was revealed, in keeping with to be related to the increased PLTP activity. previous data (Speijer et al. 1991; Francone et al. It is well established that PLTP facilitates pre 1996; Syvänne et al. 1996). β-HDL generation (Jiang et al. 1996; Dullart and In conclusion, PLTP activity was increased Van Tol 2001b) and may also directly stimulate in type 2 diabetes with CAD and those without cellular cholesterol removal by enhancing surface CAD, suggesting that the elevated activity could binding of HDL (Wolfauer et al. 1999). Indeed, be involved in the impaired cellular cholesterol cholesterol efflux from Fu5AH cells is positively removal and might accelerate atherosclerosis in correlated with plasma PLTP activity (Syvänne et patients. al. 1996; Dullart and Van Tol 2001b). In condi- tions associated with high plasma triglycerides, Acknowledgments like obesity and insulin resistance, the PLTP This work was supported by grants from «la activity, as a measure of the amount of active Direction Générale de la Recherche Scientifique, PLTP, may be elevated (Riemens et al 1999; Tunisia » and «la Coopération Francaise» (CMCU 98/F0810 and INSERM/DGRST 04/M02). Huuskonen et al. 2001). Other studies have shown a strong dependence of cellular cholesterol efflux on HDL-phospholipid composition References (Fournier et al. 1996, 1997; Jian et al. 1997). In Albers, J.J., Wolfbauer, G., Cheung, M.C., Day, J.R., Ching, A.F., Lok, S. & Tu, A.Y. (1995) Functional expression of our study, capacity of serum to promote cellular human and mouse plasma phospholipid transfer protein: cholesterol efflux was similarly related to HDL- effect of recombinant and plasma PLTP on HDL subspe- phospholipids. cies. Biochim. Biophys. Acta, 1258, 27-34. Attia, N., Durlach, V., Paul, J.-L., Soni, T., Betoulle, D. & As shown by Duell and co-workers (1991), Girard-Globa, A. (1995) Modulation of low density lipo- the decreased capacity of serum to promote cellu- protein subclasses by alimentary lipemia in control and normotriglyceridemic non insulin-dependent diabetic sub- lar cholesterol in the D+C- group could be due to, jects. Atherosclerosis, 3, 197-209. at least in part, to the phenomenon of glycation of Attia, N., Durlach, V., Roche, D., Paul, J.L., Soni, T., Zahouani, HDL apolipoproteins which is revealed to inhibit A., Landron, F., Labrousse, F., Leutenegger, M. & Girard- Globa, A. (1997) Postprandial metabolism of triglyceride- removal of intracellular cholesterol. Our results rich lipoproteins in non insulin-dependent diabetic patients indicate that PLTP activity was elevated in type 2 before and after bezafibrate treatment. Eur. J. Clin. Invest., 27, 55-63. diabetes, but when CAD coexists, it was more Autran, D., Attia, N., Dedecjus, M., Durlach, V. & Girard- accelerated and the expected positive correlation Globa, A. (2000) Postprandial reverse cholesterol transport with cholesterol efflux level was not significant, in type 2 diabetic patients: effect of a lipid lowering treat- ment. Atherosclerosis, 153, 453-460. indicating that an activated PLTP would be respon- Brites, F., Cavallero, E., De Geitere, C., Nicolaıew, N., Jacotot, sible, at least in part, for the altered cellular cho- B., Rosseneu, M., Fruchart, J.C., Wikinski, R.L. & Castro, lesterol removal, and consequently may contribute G.R. (1999) Abnormal capacity to induce cholesterol efflux and a new LpA-I pre-β particle in type 2 diabetic to the increased risk of atherosclerosis (Jiang patients. Clin. Chim. Acta, 279, 1-14. 2002; van Haperen et al. 2002; O’Brien et al. Castro, G.R. & Fielding, C.J. (1988) Early incorporation of cell-derived cholesterol into pre-β-migrating high density 2003; Schlitt et al. 2003; Cheung et al. 2006; de lipoprotein. Biochemistry, 27, 25-29. Vries et al. 2006; Vikstedt et al. 2007). More Cavallero, E., Brites, F., Delfly, B., Nicolaïew, N., Decossin, C., recent data have shown that in non-diabetic sub- De Geitere, C., Fruchart, J.C., Wikinski, R., Jacotot, B. & Castro, G. (1995) Abnormal reverse cholesterol transport jects, low PLTP activity is a marker for the pres- in controlled type II diabetic patients. Studies on fasting ence of peripheral arterial disease (Schgoer et al. and postprandial LpA-I particles. Arterioscler. Thromb. Vasc. Biol., 15, 2130-2135. 2007). Chaaba, R., Attia, N., Hammami, S., Smaoui, M., Mahjoub, S., We have shown that cholesteryl ester transfer Hammami, M. & Masmoudi, A.S. (2005a) Association of protein (CETP) activity was increased in both SNP3 polymorphism in the apolipoprotein A-V with plasma triglyceride level in Tunisian type 2 diabetes. D+C- and D+C+ patients as compared to controls Lipids Health Dis., 4, 1. (Chaaba et al. 2005b), but, no association of Chaaba, R., Hammami, S., Attia, N., Smaoui, M., Masmoudi, 136 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 137

A.S., Mahjoub, S., Ben Hamda, K. & Hammami, M. ester transfer protein (CETP) and human apolipoprotein (2005b) Association of plasma cholesteryl ester transfer A-I (apo A-I) transgenes. J. Lipid Res., 37, 1268-1277. protein activity and polymorphism with coronary artery Friedwald, W.T., Levy, R.I. & Fredrickson, D.S. (1972) Estima- disease extent in Tunisian type II diabetic patients. Clin. tion of the concentration of low-density lipoprotein choles- Biochem., 38, 373-378. terol in plasma, without use of the preparative ultracentri- Cheung, M.C., Brown, B.G., Larsen, E.K.M., Frutkin, A.D., fuge. Clin. Chem., 18, 499-502. O’Brein, K.D. & Albers, J.J. (2006) Phospholipid protein Gordon, D.J. & Rifkind, B.M. (1989) High-density lipoprotein- transfer activity is associated with inflammatory markers in the clinical implications of recent studies. N. Engl. J. Med., patients with . Biochim. Biophys. 321, 1311-3116. Acta, 1762, 131-137. Howard, B.V. (1987) Lipoprotein metabolism in diabetes Damen, J., Regts, J. & Scherphof, G. (1982) Transfer of mellitus. J. Lipid Res., 28, 613-628. [14C]phosphatidylcholine between liposomes and human Huuskonen, J., Olkkonen, V.M., Jauhiainen, M. & Ehnholm, C. plasma high density lipoprotein. Partial purification of a (2001) The impact of phospholipid transfer protein (PLTP) transfer-stimulating plasma factor using a rapid transfer on HDL metabolism. Atherosclerosis, 155, 269-281. assay. Biochim. Biophys. Acta, 712, 444-452. Ishida, B.Y., Frohlich, J. & Fielding, C.J. (1987) Prebeta- de la Llera Moya, M., Atger, V., Paul, J.L., Fournier, N., Moatti, migrating high density lipoprotein: quantitation in normal N., Giral, P., Friday, K.E. & Rothblat, G. (1994) A cell and hyperlipidemic plasma by solid phase radioimmunoas- culture system for screening human serum for ability to say following electrophoretic transfer. J. Lipid Res., 28, promote cellular cholesterol efflux: relationships compo- 778-786. nents and efflux, esterification and transfer. Arterioscler. Jian, B., de la Llera Moya, M., Royer, L., Rothblat, G.H., Thromb., 14, 1056-1065. Francone, O. & Swaney, J.B. (1997) Modification of the de Vries, R., Dallinga-Thie, G.M., Smit, A.J., Wolffenbuttel, cholesterol efflux properties of human serum by enrich- B.H., van Tol, A. & Dullart, R.P. (2006) Elevated plasma ment with phospholipids. J. Lipid Res., 38, 734-744. phospholipids transfer protein activity is a determinant of Jiang, X., Francone, O.L., Bruce, C., Milne, R., Mar, J., Walsh, carotid intima-media thickness in type 2 diabetes mellitus. A., Breslow, J.L. & Tall, A.R. (1996) Increased pre β-high Diabetologia, 49, 398-404. density lipoprotein, apolipoprotein A1, and phospholipid in Duell, P.B., Oram, J. & Bierman, E.L. (1999) Nonenzymatic mice expressing the human phospholipid transfer protein glycosylation of HDL and impaired HDL-receptor- and human apolipoprotein A1 transgenes. J. Clin. Invest., mediated cholesterol efflux. Diabetes , 40, 377-384. 96, 2373-2380. Dullart, R.P.F. & van Tol, A. (2001a) Role of phospholipid Jiang, X.C. (2002) The effect of phospholipids transfer protein transfer protein and preβ-high density lipoproteins in main- on lipoprotein metabolism and atherosclerosis. Front. taining cholesterol efflux from Fu5AH cells to plasma from Biosci., 7, d1634-6341. insulin-resistant subjects. Scand. J. Clin. Lab. Invest., 61, O’Brien, K.D., Vuletic, S., McDonald, T.O., Wolfbauer, G., 69-74. Lewis, K., Tu, A.Y., Marcovina, S., Wight, T.N., Chait, A. Dullart, R.P.F. & Van Tol, A. (2001b) Twenty four hour insulin & Albers, J.J. (2003) Cell-associated and extracellular infusion impairs the ability of plasma from healthy subjects phospholipid transfer protein in human coronary artery and Type 2 diabetic patients to promote cellular cholesterol atherosclerosis. Circulation, 108, 270-274. efflux. Atherosclerosis , 157, 49-56. Ohashi, R., Mu, H., Wang, X., Yao, Q. & Chen, C. (2005) Fielding, C.J. & Fielding, P.E. (1995) Molecular physiology of Reverse cholesterol transport and cholesterol efflux in reverse cholesterol transport. J. Lipid Res., 36, 211-228. atherosclerosis. Q. J. Med., 98, 845-856. Fournier, N., de la Llera Moya, M., Burkey, B., Swaney, J., Oram, J.F., Wolfbauer, G., Vaughan, A.M., Tang, C. & Albers, Paternini, J.J., Moatti, N., Atger, V. & Rothblat, G.H. (1996) J.J. (2003) Phospholipid transfer protein interacts with and The role HDL phospholipids in efflux of cell cholesterol to stabilizes ATP-binding cassette transporter A1 and whole serum: studies with human apo AI transgenic rats. J. enhances cholesterol efflux from cells. J. Biol. Chem., 278, Lipid Res., 37, 1704-1711. 52379-52385. Fournier, N., Paul, J.L., Atger, V., Congny, A., Soni, T., de la Riemens, S.C., Van Tol, A., Stulp, B.K. & Dullaart, R.P.F. (1999) Llera Moya, M., Rothblat, G. & Moatti, N. (1997) HDL Influence of insulin sensivity and the TaqIB cholesteryl phospholipid content and composition as a major factor ester transfer protein gene polymorphism on plasma determining cholesterol efflux capacity from Fu 5AH cells lecithin:chlesterol acyltransferase and lipid transfer activi- to human serum. Arterioscler. Thromb. Vasc. Biol., 17, ties an their response to hyperinsulinemia in non-diabetic 2685-2691. men. J. Lipid Res., 40, 1467-1474. Fournier, N., Atger, V., Cogny, A., Vedie, B., Giral, P., Simon, A., Schgoer, W., Mueller, T., Jauhiainen, M., Wehinger, A., Gander, Moatti, N. & Paul, J.L. (2001) Analysis of the relationship R., Tancevski, I., Salzmann, K., Eller, P., Ritsch, A., between triglyceridemia and HDL-phospholipid concentra- Haltmayer, M., Ehnholm, C., Patsch, J.R. & Foeger, B. tions: consequences on the efflux capacity of serum in the (2007) Low phospholipids transfer protein (PLTP) is a risk Fu5AH system. Atherosclerosis, 157, 315-323. factor for peripheral atherosclerosis. Atherosclerosis, Fournier, N., Francone, O.L., Rothblat, G., Goudouneche, D., (ahead of print). Cambillau, M., Kellner-Weibel, G., Robinet, P., Royer, L., Schlitt, A., Bickel, C., Thuma, P., Blankenberg, S., Rupprecht, Moatti, N., Simon, A. & Paul, J.L. (2003) Enhanced efflux H.J., Meyer, J. & Jiang, X.C. (2003) High plasma of cholesterol from ABCA1-expressing macrophages to phospholipid transfer protein levels as a risk factor for serum from type IV hypertriglyceridemic subjects. Athero- coronary artery disease. Arterioscler. Thromb. Vasc. Biol., sclerosis, 171, 287-293. 23, 1857-1862. Francone, O.L., Royer, L. & Haghpassand, M. (1996) Increased Speijer, H., Groener, J.E.M., Van Ramshorst, E. & Van Tol, A. preβ-HDL levels, cholesterol efflux, and LCAT-mediated (1991) Different locations of cholesteryl ester transfer esterification in mice expressing the human cholesteryl protein and phospholipid transfer protein activities in 136 N. Attia et al. Cholesterol Efflux and PLTP Activity in Diabetes with CAD 137

plasma. Atherosclerosis, 90, 159-168. T.J., Ehnholm, C., Jauhiainen, M. & Van Eck, M. (2007) Syvänne, M., Castro, G., Dengremont, C., Dengremont, C., De Macrophage phospholipids transfer protein contributes Geitere, C., Jauhiainen, M., Ehnholm, C., Michelagnoli, S., significantly to total plasma phospholipids transfer activity Franceschini, G., Kahri, J. & Taskinen, M.R. (1996) and its deficiency leads to diminished atherosclerosis lesion Cholesterol efflux from FU5AH hepatoma cells induced by development. Arterioscler. Thromb. Vasc. Biol., 27, plasma of subjects with or without coronary artery disease 578-586. and non-insulin-dependent diabetes: importance of LpA- von Eckardstein, A., Jauhiainen, M., Huang, Y., Metso, J., I:A-II particles and phospholipid transfer protein, Athero- Langer, C., Pussinen, P., Wu, S., Ehnholm, C. & Assmann, sclerosis, 127, 245-253. G. (1996) Phospholipid transfer protein mediated conver- Tall, A. & Wang, N. (2000) as a test of the sion of high density lipoproteins generated preβ1-HDL. reverse cholesterol transport hypothesis. J. Clin. Invest., Biochim. Biophys. Acta, 1301, 255-262. 106, 1205-1207. West, K.M., Ahuja, M.M., Bennett, P.H., Czyzyk, A., De Taskinen, M.R. (1990) Hyperlipideamia in diabetes. Baillieres Acosta, O.M., Fuller, J.H., Grab, B., Grabauskas, V., Clin. Endocrinol. Metab., 4, 743-775. Jarrett, R.J. & Kosaka, K. (1983) The role of circulating The Expert Committee on the Diagnosis and Classification of glucose and triglyceride concentrations and their interac- Diabetes Mellitus. (1997) Report of the Expert Committee tions with other ‘risk factors’ as determinants of arterial on the Diagnosis and Classification of Diabetes Mellitus. disease in nine diabetic population samples from the WHO Diabetes Care, 20, 1183-1197. multinational study. Diabetes Care, 6, 361-369. The sixth report of the Joint National Committee on prevention, Wolfauer, G., Albers, J.J. & Oram, J.F. (1999) Phospholipid detection, evaluation, and treatment of high blood pressure. transfer protein enhances removal of cellular cholesterol (1997) Arch. Inter. Med., 157, 2413-2446. and phospholipids by high-density lipoprotein apolipopro- van Haperen, R., van Tol, A., van Gent, T., Scheek, L., Visser, P., teins. Biochim. Biophys. Acta, 1439, 65-76. van der Kam, A., Grosveld, F. & de Crom, R. (2002) Yancey, P.G., Bortnick, A.E., Kellner-Weibel, G., de la Llera- Increased risk of atherosclerosis by elevated plasma levels Moya, M., Phillips, M.C. & Rothblat, G.H. (2003) Impor- of phospholipid transfer protein. J. Biol. Chem., 277, tance of different pathways of cellular cholesterol efflux. 48938-48943. Arterioscler. Thromb. Vasc. Biol., 23, 712-719. Vikstedt, R., Ye, D., Metso, J., Hildebrand, R.B., Van Berkel,