Tuesday June 27, 2000: Workshop Abstracts W:I 5 Genetics of Risk Factors for CVD to plaque rupture: 1) local thrombogenic substrate; 2) local flow disturbances; and 3) systemic thrombotic propensity. To study the devastating consequences of atherosclerosis we have tried to develop an animal model of plaque rupture with superimposed thrombosis. Until now, we have failed to induce rupture of "vulnerable-looking" aortic root plaques in more than 200 middle-aged (> 1 year old) apoE - / - mice exposed to extremely stressful stimuli.
TuW2:14 I Is there a mouse model of plaque rupture? C.L. Jackson. Bristol Heart Institute, University of Bristol, Bristol, UK
Objective: To determine the incidence of atherosclerotic plaque rupture at arterial branch points in apolipoprotein E knockout ( a p o E - / - ) mice. Methods: Six male and five female a p o E - / - mice were fed a diet supplemented with 21% lard and 0.15% cholesterol for up to 14 months. In animals that died, the principal branch points in the carotid arteries and aorta were removed and analysed. Sections were examined histologically, using haematuxylin and eosin. Miller's elastin stain, or Martins Scarlet Blue, and by immunocytochemistry, using an antibody directed against a-smooth muscle action to identify smooth muscle cells. Results: Four of the male mice and four of the female mice died, after 46 5= 3 weeks of feeding (range 37 to 59 weeks). Lumenal thrombus associated with atherosclerotic plaque rupture was observed in three male and all four female mice. In six of these seven mice, an atherosclerotic plaque rupture was found where the hrachiocephalic artery branches into the right common carotid and right subclavian arteries. The ruptures were characterised by fragmentation and loss of elastin in the fibrous caps of relatively small and lipid-rich plaques overlying large complex lesions, with intraplaque haemorrhage. Immunocytochemical analysis revealed loss of smooth muscle cells from raptured caps. Conclusions: These data suggest that long-term fat-feeding of apolipoprotein E knockout mice is a useful and reproducible model of atherosclerotic plaque rupture, and that these ruptures occur predominantly in the hrachiocephalic artery. ~
Visualization of the vulnerable plaque
G. Pasterkamp. Department of Cardiology, University Medical Center
Utrecht, Utrecht, The Netherlands The composition of the atherosclerotic lesion rather than the degree of stenosis is currently considered to be the most important determinant for acute clinical events. Modalities capable of characterizing the atherosclerotic lesion may help to understand its natural history and detect lesions with high risk for acute events. Grossly, three histological features of the vulnerable plaque have been reported: size of the atheroma, thickness of the fibrous cap and inflammation. Imaging techniques are currently being deployed and under development to visualize these characteristics of the vulnerable coronary plaque. Most of these diagnostic modalities have the potential to locally detect one or more of the three histologically defined features of the vulnerable plaque. The highest resolutions are achieved by catheter based techniques like IVUS and Raman spectroscopy. Except for optical coherence tomography, however, the resolution of current techniques is still too limited to discriminate the thin fibrous cap. The non invasive imaging modalities like MRI suffer from inadequate resolutions but their unlimited penetration depth and its non invasive nature are advantages. Imaging techniques that visualize the plaque locally may provide new insight into the etiology of sudden progression of atherosclerotic disease or acute events. However, due to their local applicability, it is not expected that they will have prognostic properties for the development of acute clinical syndromes that often originate from non haemodynamicaily significant lesions. Therefore, systemic markers for inflammation may have more prognostic value for the identification of patients suffering from clinical events as a result of plaque rupture.
lncrcased serum MMP-9 concentrations in patients with angiographically assessed coronary artery disease
A. Kalela I , T.A. Koivu 1, J. Kanervisto t , P. Sillanaukee 1,2, T. Lehtim~ki2, S.T. Nikkafi 1,2. 1University of Tampere, Medical School; 2Centre for
Laboratory Medicine, University Hospital of Tampere, Finland Objectives: Matrix metalloproteinases (MMPs) are upregulated in unstable atherosclerotic plaques. We examined whether MMP serum levels reflect the progression of coronary artery disease (CAD).
Methods: Serum matrix metallopmteinase-9 (MMP-9) concentrations were determined by ELISA in 61 patients (39 males, 22 females, mean age 56.5 years), who had narrowing in one or more coronary arteries assessed by coronary angiography. The control group consisted of 19 patients (9 males, 10 females, mean age 50.9 years), who had no pathological findings in coronary angiography. Results: Serum MMP-9 concentrations tended to increase in the order; controls (32.2 :t= 16.1 mg/L), 1 or 2 vessel CAD (40.4 5= 25.1 mg/L) and 3 vessel CAD (57.3 5= 39.1 mg/L) (P = 0.011, ANOVA). In a logistic regression model adjusting for known CAD risk factors, serum MMP-9 was the strongest predictor of CAD (P = 0.013). In a 10 year follow-up, the serum MMP-9 concentration, taken before coronary bypass surgery, did not predict subsequent mortality from coronary events. Conclusions: These results suggest that serum MMP-9 concentration is associated with severity of coronary narrowing and may have diagnostic value in evaluating the extent of CAD. TuW5:14 [ Lipoprotein-nssodated Phospholipnse A2 (Lp PLA2), an inflammatory marker and novel independent risk factor in the West of Scotland Coronary Prevention Study (WOSCOPS) C.J. Packard I , D. O'Reilly I , M.J. Caslake I , G. McIntosh1, C.H. Macphee2, ~ g 2 , M. Krishna3, A. Rumley I , G. Lowe I . On behalf of WOSCOPS Group; l Depts of Biochemistry and Medicine; 1Robertson
Centre, Glasgow University; 2Smith Kline Beecham Labs, Harlow, UK; 3DiaDexus, Santaclara, USA Objective: To determine the usefulness of chronic inflammation markers as predictors of risk in WOSCOPS. Methods: A nested case-control study was designed from within the WOSCOPS cohort. 580 cases with CHD were age and smoking matched with 1160 controls. C Reactive Protein (CRP), Lp PLA2 mass, fibrinogen (Fib) and white cell count (WCC) were measured at baseline or on frozen, stored samples. Relationship to risk was tested in Cox proportional hazards models. Risk ratios were estimated for quintiles (Q) of each variable. Results: All four markers were strong predictors of risk in univariate analysis with about a 2 fold increase in risk comparing subjects in the highest vs lowest qulntile. CRP correlated with Fib, WCC, body mass index, HDL and plasma triglyceride and in multivariate analysis its predictive capacity (like that of Fib and WCC) was much reduced. Lp PLA2, was not correlated with other inflammatory markers and remained a strong independent risk factor in multivariate analysis. For Lp PLA2, risks in Q 2-4 relative to Q1 were 1.26 (CI 0.83-1.92), 1.58 (1.04-2.4), 1.73 (1.16-2.60) and 1.66 (1.10-2.50) adjusting all other risk factors. Conclusions: Chronic inflammation is a strong determinant of risk in asymptomatic, moderately hypemholesterolemic men. Lp PLA2, an entity distinct from Group H secretory PLA2, is identified as a new, independent marker of CHD risk.
W:15 GENETICS OF RISK FACTORS FOR CVD
Genomic searches for genes that influence risk factors for CVD
J.E. Hixson. Dept. of Genetics, S. W. Found.for Biomed. Res., San Antonio,
TX, USA Objective: To identify quantitative trait loci (QTLs) that influence CVD risk factors using genomic screens in families. Methods: Our general approach is to perform genome-wide searches in large extended families that are not selected according to any particular disease. The genetic markers that are used for genomic searches are random microsatellite markers distributed throughout the human chromosomes at approximately 10 cM intervals. These markers are used for linkage analysis with variance component methods to identify chromosomal regions containing QTLs for CVD risk factors. Results: We have conducted a genomic search in 480 family members from 10 large extended pedigrees of Mexican Americans in San Antonio. Our first published finding was the identification of a major QTL on chromosome 2 that influences serum levels of leptin hormone, an important risk factor for obesity. We have also identified QTLs on chromosomes 3 and 4 that influence LDL size class, an important CVD risk factor. In addition to lipid risk factors,
XIlth International Symposium on Atherosclerosis, Stockholm, Sweden, June 25-29, 2000
Tuesday June 27, 2000: Workshop Abstracts W:I5 Genetics of Risk Factors for CVD
we are measuring levels of gene products involved in atherogenesis in the arterial wall. For example, we have found strong evidence for genetic control of serum levels of soluble P-se&tin (h’ = 0.70), and have detected major QTLs on chromosome 15 (LOD = 3.8) and chromosome 12 (LOD = 2.6). Conclusions: Genomic searches in families are a powerful strategy to identify new QTLs that influence CVD risk factors.
Genetic factors and the response of CVD risk factors to regular exercise
C. Bouchard. Pennington Biomedical Research Center; Baton Rouge, IA, USA Objective: To review the evidence for familial aggregation and the contribu-
tion of specific genes in the response of CVD risk factors to regular physical activity. The 1996 USA Surgeon General’s Report on physical activity and health reviews several prospective studies concerning the relationship between levels of physical activity, CVD outcomes and CVD mortality rates. Collectively, these studies indicate that physically active adults are, on the average, less prone to CVD and have lower death rates from CVD than sedentary people. There are considerable individual differences in the changes observed in lipids and lipoproteins, blood pressure and other CVD risk factors as a result of exposure to exercise programs. A handful of twin studies and more recently the HERITAGE Family Study have shown that there is strong familial aggregation for the response pattern to regular exercise. The heritability for the changes in maximal oxygen uptake to a standardized exercise training program attained about 50 percent in HERITAGE. A genomic scan performed has revealed that several QTLs are linked to the responsiveness to regular exercise for cadiorespiratory endurance. However, thus far, candidate gene studies have evidenced only weak associations with the changes in CVD risk factors with regular exercise. The genetic dissection of the familial component of the response to regular exercise will be a complex undertaking requiring a variety of genomic and expression technologies.
Gene environment interaction in determining ischaemic heart disease
S.E. Hum hries’, P.J. Talmud’, J. Coopeta, D.M. Waterworth’,I.N.M. Day’,
Linkage of blood pressure to a locus on chromosome dutch dyslipidemic famiiies
H. Allaveet, T.W.A. de Bruin’, R.M. Cantor’, B.E. Aouizeratt , A.J. Lusis’, J.I. Rotters. ‘Dept. of Human Genetics, UCLA; 3Div. of Medical Genetics, Steven Spielberg Research Center Cedars-Sinai Medical Center; Los Angeles, USA; 2Dept. of Medicine, Academic Hospital, Maastricht, The Netherlands
Genes contributing to the common forms of essential hypertension and blood pressure (BP) variation are largely unknown. This may result from the underlying genetic heterogeneity of this disorder. One appmach to reduce such heterogeneity is to conduct gene finding efforts in families ascertained for common metabolic syndromes with associated hypertension. Familial combined hyperlipidemia (FCHL), a metabolic syndrome associated with insulin resistance, central obesity, and an increased frequency of hypertension, is such a disorder. Insulin resistance, defined by fasting insulin levels and more direct measures, is both associated with hypertension in cross-sectional studies and predictive in prospective studies. Furthermore, insulin resistance and FCHL are also accompanied by increased free fatty acid (FFA) levels. In the present study, we analyzed a 10 CM genome-wide scan in 18 Dutch FCHL pedigrees (N = 240) to search for genes contributing to BP in this metabolic syndrome. A multipoint genome scan of systolic (S) BP and diastolic (D) BP identified a region on chromosome 4 exhibiting a LOD score of 3.9 with SBP (peak marker D452639). Interestingly, FFA levels mapped to this region, with a LOD score of 2.4. Two-point linkage with markers under the peak also yielded evidence for linkage of SBP (P <: 0.0008) and FFA (P < 0.009) to this locus, supporting the multipoint results. Alpha adducin, a gene involved in renal sodium handling, resides within this locus and has been associated with elevated blood pressure in both human populations and in animal models of hypertension. However, there is no evidence for an association between two intragenic polymorphisms within a-adducin and SBP in our Dutch population. In conclusion, this genome scan for BP has identified a chromosomal region harboring a potentially novel gene that contributes to hypertension associated with insulin resistant dyslipidemia.
A quantitative trait locus for vascular cellular adhesion molecule-l levels on chromosome 19
MC. Mahaney’, L. Almasyi, A.G. Comuzzie’, S.A. Cole’, J.E. Hixson’, J.W. MacCluer’ , M.P. Stem2, J. Blangero’. ‘Southwest Foundation for
‘British Heart Foundation Centre for Cardiovascular Genetics, LzYzz&Royal Free and University College London Medical School, London WCIE 6JJ; 2MRC Epidemiology and Medical Care Unit, London ECIM SBQ, UK
Biomedical Research; ‘University of Texas Health Science Center; San Antonio, Texas, USA
In understanding the impact of risk of thrombosis and Ischaemic Heart Disease
influencing serum vascular cellular adhesion molecule- 1 (VCAM- 1) levels in Mexican Americans from San Antonio, Texas. Methods: We assayed serum levels of VCAM-1 in 471 Mexican Americans from 10 extended pedigrees using a commercially available ELISA kit. Initial statistical genetic analyses and a two-point, 20 CM whole genome linkage screen was accomplished using a maximum-likelihood based, variante-decomposition approach. We followed these analyses with a multipoint, 15 CM, whole genome linkage screen. Results: Our initial analyses showed that variation in VCAM-1 levels was influenced by additive genes (h2 = 0.26). sex, diabetes, smoking, and menopause and provided suggestive evidence for QTLs on chromosomes 19 (LOD = 2.36) and 14 (LOD = 2.02). The 15 CM multipoint linkage screen yielded a maximum multipoint LOD score of 3.25 (p = 0.000122) on chromosome 19p. The 95% confidence interval localizes this QTL to 19~13.3 and subsumes a region containing excellent candidate loci for vascular biology; including loci for the mucosal addressin cell adhesion molecule-l (MAdCAM-I), the thromboxane A2 receptor (TBXA2R). and intercellular adhesion molecules-l and-3 ([CAM-I and ICAM-3). Conclusions: A small-but-significant proportion of the phenotypic variance in normal serum levels of VCAM-1 is due the additive effects of genes. A QTL responsible for most of the additive genetic effect on VCAM-1 levels in this population is located on chromosome 19~13.3.
(IHD), the modifying effects of different environmental factors experienced by individuals on the predisposition that they have inherited is particularly important. IHD risk has been determined in the second Northwick Park Heart Study, at ten candidate gene loci; apolipoprotein (ape) CIII (C3238G [SstI]) and C-482T [insulin responsive element, IRE]), apoAIV (Thr347Ser), apoE (E2, E3, E4), apoB (C7672T [XbaI]), Lipoprotein Lipase (Ser447Stop). beta-fibrinogen (G-455A). factor VII (Arg353Gln). ACE (I/D), eNOS (intron 4) and stromelysin-1 (5A/6A). DNA was available on 2743 middle-aged men, free of II-ID at baseline, recruited for prospective cardiovascular surveillance. Carriers of the apoAIV gene Ser347 allele had a Relative Risk (RR) of 1.57 (95% CI: 1.07-2.31, p = 0.04) compared to Thr347 homozygous men, while stromelysin 6A6A homozygous men had a RR of 1.91 (1.1 l-3.28, p = 0.02) compared with those with genotype 5A5A. Overall, men who were current smokers had a 2.20 (1.52-3.17, p < 0.0001) fold higher risk of IHD, and there was evidence for interaction between smoking and genotype in modulating risk at the stromelysin (p = 0.09), apoE (p = 0.02) and apoCII1 (p = 0.002) loci. In particular, smoking was not associated with increased risk in the apoE3E3 group, but markedly augmented risk in apoE carriers (RR = 2.22 (1.33-3.72) in smokers compared with 0.58 (0.32-1.05) in non-smokers. Estimates of all of the genotype-associated IHD risk effects remained essentially unchanged after adjustment for the classical risk factors, suggesting that risk is not mediated through effects on plasma levels of measured lipids or clotting factors. Since men with the stromelysin genotype 5A5A and those who carry the apoE allele each represent 25% of the general population, this provides a strong argument for smoking avoidance in these individuals.
Objective: To detect, characterize and localize quantitative trait loci (QTLs)
I TuW6 15
Acute coronary event risk-increasing polymorphisms adrenergic receptor and serum paraoxonase genes. demonstration of a gene-gene instruction
T.-P. Tuomainen’, T.A. Lakkai, A. Sna i? R Malir?, T. Lehtimlki3, p,: M. Koult?, M. Scheinin*, J.T. Salonen . ‘Unrversrty of Kuopio, Kuopio;
2University of Turku, Turku; 3 University of Tampere, Tampere, Finland Objective:
We hypothesised that multiple gene polymorphisms which pm-
XIIth International Symposium on Atherosclerosis, Stockholm, Sweden, June 25-29, 2ooO