ApoC-III gene polymorphisms and risk of coronary artery disease

• DOI: 10.1194/jlr.m200145-jlr200
• Publication Date: 2002-11-27T14:42:33Z
• Authors: Oliviero Olivieri; Chiara Stranieri; Antonella Bassi; Barbara Zaia; Domenico Girelli; Francesca Pizzolo; Elisabetta Trabetti; Suzanne Cheng; Michael A. Grow; Pier Franco Pignatti; Roberto Corrocher
• Keywords: Adult; Male; insulin; Genotype; QD415-436; Coronary Artery Disease; Biochemistry; lipids; 03 medical and health sciences; Endocrinology; 0302 clinical medicine; Odds Ratio; risk factors; Humans; Insulin; Genetic Predisposition to Disease; Apolipoproteins C; Apolipoproteins B; Insulin; Lipids; Risk factors;; Apolipoprotein C-III; Polymorphism, Genetic; Apolipoprotein A-I; Cell Biology; Middle Aged; Female; Biomarkers

Several polymorphisms in the apolipoprotein C-III (apoC-III) gene have been associated with hypertriglyceridemia, but link coronary artery disease risk is still controversial. In particular, apoC-III promoter sequence variants insulin responsive element (IRE), constitutively resistant to downregulation by insulin, never investigated this connection. We studied a total of 800 patients, 549 whom had angiographically documented atherosclerosis, whereas 251 normal arteriograms. measured plasma lipids, apoA-I, apoB, and assessed three gene, namely, T-455C IRE region, C1100T exon 3, Sst1 polymorphic site (S1/S2) 3′ untranslated region. Each variant influenced triglyceride levels, only (in homozygosity) S2 alleles levels. (CAD) 18.6% were homozygous for −455C compared 9.2% CAD-free group (P < 0.001).In logistic regression models, homozygosity was significantly increased CAD (OR = 2.5 2.18 unadjusted adjusted respectively) suggesting that it represents an independent genetic susceptibility factor CAD. 0.001). Investigators long disputed whether elevated serum (TG) levels are (1Bloomfield Rubins H. The trouble triglycerides.Arch. Intern. Med. 2000; 160: 1903-1904Google Scholar). A major reason controversy stems from heterogeneity factors determination (TGs carried virtually all lipoproteins) high variance collinearity TGs other recognized Scholar, 2Hodis H.N. Triglyceride-rich lipoprotein remnant particles atherosclerosis (Editorial).Circulation. 1999; 99: 2852-2854Google Alternative, more reliable, markers TG metabolism therefore proposed Perhaps most important among them (apoC-III), 79 amino acid protein synthesized liver intestine, which essential constituent circulating rich triacylglycerol, i.e., chylomicrons VLDL (3Jong M.C. Hofker M.H. Havekes L.M. Role apo Cs metabolism. Functional differences between apoC1, apoC2, apoC3.Arterioscler. Thromb. Vasc. Biol. 19: 472-484Google results large clinical studies indicated better predictor development progression than traditionally (4Blankenhorn D.H. Alaupovic P. Wickham E. Chin H.P. Azen S.P. Prediction angiographic change native human arteries aortocoronary bypass grafts. Lipid nonlipid factors.Circulation. 1990; 81: 470-476Google 5Hodis Mack W.J. Pogoda J.M. La Bree L. Hemphill L.C. Kramsch D.M. Blankenhorn Triglyceride cholesterol-rich lipoproteins differential effect on mild/moderate severe lesion as quantitative angiography controlled trial lovastatin.Circulation. 1994; 90: 42-49Google 6Mack Krauss R.M. Hodis Lipoprotein subclasses monitored study (MARS). Treatment effects relation progression.Arterioscler. 1996; 16: 697-704Google 7Luc G. Fievet C. Arveiler D. Evans A.E. Bard Cambien F. Fruchart J.C. Ducimetiere Apolipoproteins et al. C–III E B- non-apo B-containing two populations at contrasting myocardial infarction: ECTIM study.J. Res. 37: 508-517Google 8Thompson G.R. Angiographic evidence role triglyceride-rich disease.Eur. Heart J. 1998; H31-H36Google 9Sacks F.M. Moye L.A. Cole T.G. Sussex B. Stampfer M.J. Pfeffer Braunwald VLDL, apolipoproteins B, CIII, E, recurrent events cholesterol (CARE) trial.Circulation. 102: 1886-1892Google ApoC-III inhibits hydrolysis TG-rich lipase their apoE–mediated hepatic uptake (10McConathy Gesquiere Bass Tartar A. Wang C.S. Inhibition activity synthetic peptides C–III.J. 1992; 33: 995-1003Google 11Ginsberg Le N.A. Goldberg I.J. Gibson Rubinstein Wang-Iverson Norum N. Brown W.V. Apolipoprotein B subjects deficiency CIII AI. Evidence catabolism vivo.J. Clin. Invest. 1986; 78: 1287-1295Google vitro transgenic animal demonstrated overexpression causes delayed clearance plasma, resulting overt hypertriglyceridemia has mapped chromosome 11 several possible atherosclerosis-related "intermediate phenotype" (12Talmud P.J. Humphries S.E. variation dyslipidemia.Curr. Opin. Lipidol. 1997; 8: 154-158Google Biallelic described 5′ region (five strong linkage disequilibrium one another: T-455C, C-482T, T-625 deletion, G-630A, C-641A) (13Dammeman M. Sandkuijl Halaas J.L. Chung W. Breslow An aplotype protective against specified polymorphisms.Proc. Natl. Acad. Sci. USA. 1993; 4562-4566Google Scholar), 3 (C1100T) (the so called Sstl S1/S2) over last decade, latter polymorphism, also extensively studied, consistently reported be (14Ordovas Civeira Genest Jr., Craig S. Robbins A.H. Meade T. Pocovi Frossard P.M. Masharani U. Wilson P.W. Salem D.N. Ward R.H. Schaefer E.J. Restriction fragment length A-I, C–III, A-IV locus. Relationships apolipoproteins, premature disease.Atherosclerosis. 1991; 87: 75-86Google 15Surgucho Page Patsch Boerwinkle Polimorphic flanking regions hypertriglyceridemia.Arterioscler. 941-947Google Despite expected implications terms cardiovascular morbidity, association allelic controversial 16Kee Amouyel Fumeron Cambou J.P. Dallongeville Lack variations AI-CIII-AIV cluster infarction sample European male: study.Atherosclerosis. 145: 187-195Google 17Russo Meigs J.B. Cupples Demissie Otvos J.D. Lahoz Cucinotta Couture Mallory Ordovas Association Sst-I polymorphism APOC3 locus lipids subclass profiles risk: Framingham offspring 2001; 158: 173-181Google Less information available regarding linked (18Peacock R.E. Hamsten Johansson Nilsson-Ehle genotypes AI-CIII-AIV, loci density subclasses.Clin. Genet. 46: 273-282Google 19Hegele R.A. Small complex diseases: review regulatory dyslipoproteinemia atherosclerosis.Clin. Biochem. 30: 183-188Google This relatively limited interest may particularly surprising if considers expression concerning (IRE) interaction (20Chen Li Leff Transcriptional regulation diabetic mice: correlation changes levels.J. 35: 1918-1924Google animals cultured cells, transcriptionally downregulated 1995, showed that, unlike wild-type promoter, containing positions –455 –482 remains active 108-fold range concentrations inasmuch sequences reduced affinity nuclear transcription mediating response (21Li W.W. Dammeman Smith Metzger Common abolishes contribute hypertriglyceridemia.J. 1995; 96: 2601-2605Google first example "insulin resistance" level Such considerations prompt speculation links these hypothetical related "intermediate" phenotypes, characterized synthesis apoC-III- lipoproteins, turn imply light elements, we designed case-control patients or without evaluate: i) and/or ii) distribution criteria selection population already detail elsewhere (22Girelli Russo Ferraresi Olivieri O. Pinotti Friso Manzato Mazzucco Bernardi Corrocher R. Polymorphisms VII disease.N. Engl. 343: 774-780Google brief, unrelated adult both sexes who recruited those referred Institute Cardiovascular Surgery Cardiovascular-Hypertension Unit Department Internal Medicine University Verona Italy. Of documented, severe, often multivessel candidates coronary-artery grafting. severity evaluated counting number epicardial (left anterior descending, circumflex, right) affected ⩾1 significant stenosis (⩾50%). As control group, considered (CAD-free), examined reasons 90% cases valvular heart disease; remaining miscellany conditions including atypical chest pain uncertain origin, congenital disease, etc.). controls required no angiogram, history nor vascular beds. Since primary aim our provide objective clear-cut definition atherosclerotic phenotype, nonsignificant (<50%) not included study. angiograms cardiologists unaware All participants came northern Italy similar socioeconomic ethnic backgrounds. At time blood sampling, complete pharmacological history, presence absence such smoking, hypertension, diabetes mellitus, obtained patients. Patients taking statins fibrates (n 266) excluded genotype-phenotype because lowering drugs (5Hodis 23Packard C.J. Overview fenofibrate.Eur. A62-A65Google approved institutional boards. Either written oral informed consent after full explanation Samples venous drawn each subject overnight fast within 10 days procedures. Serum common biochemical parameters determined previously Insulin immunometric sandwich assay (Immulite 2000 Insulin) Diagnostic Products Corporation, Los Angeles, CA; intra- interassay CVs method <5%. ApoA-I apoB commercially nephelometric immunoassays; antisera, calibrators, BNII nephelometer Dade Behring, Marburg, Germany. Intraassay CV calculated replicates duplicates days. Imprecision manufacturer specifications, intraassay 2.1 1.6% 3.2 2.36 apoA-I respectively. fully automated turbidimetric immunoassay. reagents Wako Pure Chemical Industries (Osaka, Japan) procedure recommended manifacturer implemented RXL Dimension Analyzer (Dade International Inc. Newark, DE). Sample values interpolation spectrophotometric wavelengths measurements logit, five-points, calibration curve, covering 0.0–20.0 mg/dl; 20–30 mg/dl, smaller volume automatically rerun instrument, >30 diluted manually. pools sera low, medium apoC-III; 1.84, 2.02, 1.98% 4.4, 3.4, 2.29% medium, concentration, Mutation analysis (as well routine analysis) conducted blinded subject. Three mapping (T-455C), coding (C1100T), (S1/S2), (Fig. 1). Genomic DNA prepared whole samples phenol-chloroform extraction then used according recently multilocus genotyping protocol (24Cheng Grow M.A. Pallaud Klitz Erlich H.A. Visvikis Chen J.J. Pullinger C.R. Malloy Siest Kane candidate risk.Genome 9: 936-949Google Briefly, amplified 33 cycle Multiplex Polymerase Chain Reactions (32 ng genomic each) PCR products hybridized array immobilized oligonucleotide probes. colorimetric detection based upon streptavidin-horseradish peroxidase method. computations performed using SPSS 10.0 statistical package (SPSS Inc., Chicago, IL). Distributions continuous variables groups expressed means ± SD. Logarithmic transformation skewed variables, TG, computed corresponding log-transformed values, although, sake simplicity clarity, crude data Results. Statistical significance Student's t-test, tested one-way ANOVA age sex (General Linear Model procedure). Qualitative analyzed chi-square test. Pearson coefficient variables. patient (with CAD), genotype frequencies predicted basis Hardy-Weinberg equilibrium. Intragenic haplotypes multiple estimated EH program (25Terwilliger Ott Handbook linkage. John Hopkins Press, Baltimore, MD1994: 188-198Google Scholar) detect pairwise (D′). lipid different ANOVA, Tukey post hoc multivariate comparison (ANOVA). For polymorphisms, computation considering less frequent allele recessive (homozygous vs. patients). To assess extent various odds ratios 95% confidence intervals logistic-regression analysis. separate genotype, dummy used, reference group. Adjustment patients' conventional (age, gender, smoking status, cholesterol, triglycerides, apoB) done covariates second set models. characteristics summarized Table 1. higher There (Table whole, statistically correlated (R 0.40, P 0.001), LDL 0.238, HDL (HDL-C) −0.08, 0.05) and, strongly, 0.68, 0.001).TABLE 1Characteristics controlsaPlus-minus SD.ParametersCAD 549)CAD-free 251)P valueAge (years)60.4 9.457.6 12.6<0.01Male (%)81.866.9<0.001BMI (kg/height2)bAge- sex-adjusted values.26.5 3.325.3 3.4<0.001CholesterolbAge- values.Total (mmol/l)5.83 1.125.51 1.050.001LDL (mmol/l)3. 88 0.983.53 0.93<0.001HDL (mmol/l)1.20 0.321.42 0.43<0.001Triglycerides (mmol/l)bAge- values.2.01 1.131.50 0.71<0.001ApoA-I (g/l)bAge- values.1.31 0.241.42 0.31<0.001ApoB values.1.22 0.301.06 0.25<0.001ApoC-III (mg/dl)bAge- values.12.31 4.410.7 3.25<0.001Insulin (μIU/ml)bAge- values.14.64 7.715.75 12.2NSSmoking (%)69.741.4<0.001Hypertension (%)58.330.8<0.001Diabetes (%)21.913.3<0.01Patients lipid-lowering drugs266—Statistical unpaired t-test χ2 test when appropriate. value <0.05.To convert milligrams per deciliter, divide 0.02586. triglycerides 0.01129.a Plus-minus SD.b Age- values. Open table new tab <0.05. 0.01129. Genotype 2. equilibrium No individuals found. S1/S2 contrast, frequency observed free (18.6 9.2%, 0.001; 2).TABLE 2ApoC-III controlsGenotype (% Patients)CAD-free 251)CAD 549)Chi- SquareP valueT-455C−455 TT(%) 43.8 35.3−455 TC (%) 47 46.1−455 CC(%) 9.2 18.613.070.001C1100T1100 CC 50.2 54.31100 CT 43 37.31100 TT 6.8 8.42.529NSS1/S2S1/S1 85.2 82.4S1/S2 14.8 17.60.95NS Strong (D' 0.98, D' 0.97, observed, 0.13) It should noted out 125 homozygotes −455 variant, there 55 (44%) 13 (10.4%) 1100TT genotype. order estimate impact parameters, entire population. Due known 266, patients) Results apoC-III, 534), 3. Different (data shown). Homozygous carriers genotypes, evident assuming transmission model 3). Homozygotes heterozygous (whether separately single 3).TABLE 3Levels TGs, population, (by ANOVA)ApoC-III GenotypesNo. 534)ApoC-III (mg/dl)Triglycerides (mmol/l)Insulin (μIU/ml)−455TT20611.47 3.91.79 0.94aStatistically allele.15.7 11−455TC25211.30 3.71.73 0.86aStatistically allele.14.6 8.8−455CC7612.65 5.12.12 1.4014.8 7.2−455TT −455TC45811.38 3.8aStatistically allele.1.76 0.89aStatistically allele.15.1 9.9−455CC7612.65 7.21100CC27411.3 3.61.73 0.85aStatistically allele.15.3 10.61100CT21811.8 4.51.85 1.0514.9 8.71100TT4212.1 4.52.13 1.4014.1 5.71100CC 1100CT49211.5 4.01.79 9.81100TT4212.1 5.7S1/S143811.3 3.8bStatistically patients.1.75 0.92bStatistically patients.14.9 9.9S1/S29612.8 4.82.10 1.2215.6 7.7a Statistically allele.b raising confined homozygotes), being substantial fasting due evident. Similarly, none (total, LDL, HDL-C, any relevant performed. Crude shown 4. greatest conferred 2-fold probability disease. main produced result, indicating polymorp