Pathophysiology/Complications ORIGINAL ARTICLE Atorvastatin Decreases Apolipoprotein C-III in Apolipoprotein B–Containing Lipoprotein and HDL in Type 2 Diabetes A potential mechanism to lower plasma triglycerides 1 GEESJE M. DALLINGA-THIE, PHD ON BEHALF OF THE DIABETES ATORVASTATIN ypertriglyceridemia, low levels of 2 INGRID I.L. BERK-PLANKEN, MD, PHD LIPID INTERVENTION (DALI) STUDY HDL cholesterol, and preponder- 2 AART H. BOOTSMA, MD, PHD GROUP* 2,3,4 ance of small, dense LDL particles ANS ANSEN PHD H H J , are risk factors for cardiovascular disease (CVD) (1,2) and are part of the athero- genic lipoprotein profile often present in patients with type 2 diabetes. The initia- tion of the atherogenic lipoprotein profile OBJECTIVE — Apolipoprotein (apo)C-III is a constituent of HDL (HDL apoC-III) and of remains to be established. Kissebah et al. apoB-containing lipoproteins (LpB:C-III). It slows the clearance of triglyceride-rich lipoproteins (TRLs) by inhibition of the activity of the enzyme lipoprotein lipase (LPL) and by interference (3) demonstrated that in type 2 diabetic with lipoprotein binding to cell-surface receptors. Elevated plasma LpB:C-III is an independent subjects, apolipoprotein (apo)B and risk factor for cardiovascular disease. We studied the effect of atorvastatin on plasma LpB:C-III VLDL triglyceride secretion was en- and HDL apoC-III. hanced, but did not necessarily lead to hypertriglyceridemia. In Pima Indians, RESEARCH DESIGN AND METHODS — We studied the effect of 30 weeks’ treatment hypertriglyceridemia may develop with- with 10 and 80 mg atorvastatin on plasma apoC-III levels in a randomized, double-blind, out increased triglyceride production due placebo-controlled trial involving 217 patients with type 2 diabetes and fasting plasma triglyc- to a decreased VLDL triglyceride clear- erides between 1.5 and 6.0 mmol/l. ance capacity (4). Wilson et al. (5) RESULTS — Baseline levels of total plasma apoC-III, HDL apoC-III, and LpB:C-III were showed that in a pedigree with type 2 41.5 Ϯ 10.0, 17.7 Ϯ 5.5, and 23.8 Ϯ 7.7 mg/l, respectively. Plasma apoC-III was strongly diabetes, hypertriglyceridemia only de- correlated with plasma triglycerides (r ϭ 0.74, P Ͻ 0.001). Atorvastatin 10- and 80-mg treat- veloped in subjects with decreased li- ment significantly decreased plasma apoC-III (atorvastatin 10 mg, 21%, and 80 mg, 27%), HDL poprotein lipase (LPL) activity due to a apoC-III (atorvastatin 10 mg, 22%, and 80 mg, 28%) and LpB:C-III (atorvastatin 10 mg, 23%, genetic defect. Therefore, delayed effi- and 80 mg, 28%; all P Ͻ 0.001). The decrease in plasma apoC-III, mainly in LpB:C-III, strongly ciency of clearance of triglyceride-rich li- correlated with a decrease in triglycerides (atorvastatin 10 mg, r ϭ 0.70, and 80 mg, r ϭ 0.78; poproteins (TRLs) may be a major cause Ͻ P 0.001). Atorvastatin treatment also leads to a reduction in the HDL apoC-III–to–HDL of the development of hypertriglyceride- cholesterol and HDL apoC-III–to–apoA-I ratios, indicating a change in the number of apoC-III mia in type 2 diabetes. The clearance of per HDL particle (atorvastatin 10 mg, Ϫ21%, and 80 mg, Ϫ31%; P Ͻ 0.001). TRLs in situ is dependent on the presence CONCLUSIONS — Atorvastatin treatment resulted in a significant dose-dependent reduc- of stimulatory and inhibitory factors. In tion in plasma apoC-III, HDL apoC-III, and LpB:C-III levels in patients with type 2 diabetes. vitro studies (6,7) have implicated apoC- These data indicate a potentially important antiatherogenic effect of statin treatment and may III as a noncompetitive inhibitor of LPL. explain (part of) the triglyceride-lowering effect of atorvastatin. ApoC-III may also attenuate the clearance of TRL remnants by interfering with the Diabetes Care 27:1358–1364, 2004 hepatic lipoprotein receptors (8,9). Stud- ies with genetically modified animals sup- ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● port the hypothesis that apoC-III affects the metabolism of TRLs. Transgenic mice 1 From the Department of Internal Medicine, University Medical Center Utrecht, Utrecht, the Netherlands; overexpressing the human apoC-III gene the 2Department of Internal Medicine, Erasmus Medical Center, Rotterdam, the Netherlands; the 3Depart- ment of Biochemistry, Erasmus Medical Center, Rotterdam, the Netherlands; and the 4Department of developed severe hypertriglyceridemia Clinical Chemistry, Erasmus Medical Center, Rotterdam, the Netherlands. (10) due to impaired catabolism of TRLs Address correspondence and reprint requests to Dr. G.M. Dallinga-Thie, Laboratory of Vascular Medicine (11,12), whereas mice lacking apoC-III and Metabolism, Room Bd 277, Department of Internal Medicine, Erasmus MC, Dr Molewaterplein 40, 3015 showed a very efficient clearance of VLDL GD Rotterdam, P.O. Box 2040, 3000CA Rotterdam, Netherlands. E-mail: [email protected]. Received for publication 7 October 2003 and accepted in revised form 16 March 2004. particles and had enhanced selective up- *An appendix at the end of this article lists the members of the DALI study group. take of VLDL cholesterol esters (13,14). Abbreviations: apo, apolipoprotein; CVD, cardiovascular disease; DALI, Diabetes Atorvastatin Lipid In healthy control subjects and hy- Intervention; HDL apoC-III, apoC-III in HDL; LpB:C-III, apoC-III in apoB-containing lipoprotein; LPL, perlipidemic patients, plasma triglycer- lipoprotein lipase; TRL, triglyceride-rich lipoprotein. A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion ides and apoC-III concentration are factors for many substances. strongly correlated (15,16). Patients with © 2004 by the American Diabetes Association. elevated levels of plasma triglycerides had 1358 DIABETES CARE, VOLUME 27, NUMBER 6, JUNE 2004 Dallinga-Thie and Associates Table 1 —Baseline characteristics RESEARCH DESIGN AND METHODS — This study is part of the Atorvastatin Diabetes Atorvastatin Lipid Intervention (DALI) study. DALI is a randomized, Placebo 10 mg 80 mg double-blind, placebo-controlled, multi- n 71 72 70 center study conducted in the Nether- Male sex (%) 46.5 61.1 54.3 lands (30). Briefly, 217 patients, aged Ϯ Ϯ Ϯ 45–75 years, with type 2 diabetes for at Age (years) 59 8608608 Յ BMI (kg/m2) 32.2 Ϯ 6.1 30.0 Ϯ 3.8 30.4 Ϯ 4.5 least 1 year and an HbA1c 10%, partic- Fasting glucose (mmol/l) 10.53 Ϯ 3.66 10.52 Ϯ 3007 10.66 Ϯ 2.98 ipating in the DALI study were random- Ϯ Ϯ Ϯ ized to placebo, atorvastatin 10 mg, or HbA1c (%) 8.34 1.14 8.26 1.16 8.42 1.12 LPL (units/l) 126.8 Ϯ 52.4 131.9 Ϯ 52.9 129.5 Ϯ 46.4 atorvastatin 80 mg for 30 weeks to eval- Triglycerides (mmol/l) 2.63 Ϯ 0.91 2.55 Ϯ 0.88 2.84 Ϯ 1.12 uate the effect of treatment on lipid me- Total cholesterol (mmol/l) 6.04 Ϯ 0.81 5.90 Ϯ 0.90 6.02 Ϯ 0.90 tabolism. Type 2 diabetes was defined LDL cholesterol (mmol/l) 3.80 Ϯ 0.79 3.70 Ϯ 0.91 3.74 Ϯ 0.90 according to the American Diabetes Asso- HDL cholesterol (mmol/l) 1.05 Ϯ 0.21 1.05 Ϯ 0.26 1.04 Ϯ 0.24 ciation classification (33). Inclusion crite- ApoA-I (g/l) 1.41 Ϯ 0.19 1.39 Ϯ 0.20 1.40 Ϯ 0.21 ria were fasting plasma triglycerides ApoB (mg/100 ml) 1.28 Ϯ 0.19 1.22 Ϯ 0.20 1.24 Ϯ 0.24 between 1.5 and 6.0 mmol/l, total choles- terol between 4.0 and 8.0 mmol/l, and no Data are means Ϯ SD. history of CVD. Patients were recruited in Leiden, Rotterdam, and Utrecht. The eth- increased plasma levels of apoC-III, with a ments aimed at affecting apoC-III levels ics committees of the participating centers fourfold larger proportion of apoC-III areofgreatinterest.Statinsmaylowerplas- approved the study protocol, and all pro- present in VLDL-sized particles com- ma apoC-III. In 27 patients with primary cedures followed were in accordance with pared with normolipidemic individuals hypertriglyceridemia, 20–40 mg atorva- institutional guidelines. The primary end (17–20). In contrast, patients with a ge- statin for 4 weeks reduced plasma apoC- point was the percentage decrease in netic deficiency in apoC-III had low levels III by 18–30% (28). In 305 patients with plasma triglyceride levels after atorvasta- of plasma VLDL because of an enhanced primary hypercholesterolemia (29), both tin treatment, as described elsewhere fractional catabolic rate and a rapid con- atorvastatin and pravastatin reduced LpB: (30). The secondary end point was version of VLDL into intermediate- C-III by ϳ30%. In contrast, in patients changes in specialized lipid parameters. density lipoprotein and LDL (21). ApoC- with coronary artery bypass graft, lova- Written informed consent was obtained III in apoB-containing lipoproteins (LpB: statin therapy lowered the concentrations from all subjects. C-III) mainly reflects apoC-III present on of apoB-containing lipoproteins (LDL), triglyceride-rich particles and may di- but not plasma apoC-III or LpB:C-III Analytical methods rectly affect LPL activity and receptor (19). Blood samples were drawn after a fasting binding. HDL functions as a sink for sur- Atorvastatin is a powerful HMG-CoA period of 12 h at baseline and after 30 plus apoC-III in plasma, and a high level reductase inhibitor proven to effectively weeks’ treatment at the end of the study. may indirectly affect triglyceride clear- reduce total plasma cholesterol and tri- Plasma was prepared by immediate cen- ance (22). trifugation, and samples were stored at glyceride levels.