In the homozygous form, amino acid changing mutations

Transcription

1 Lipoprotein Lipase Mutations, Plasma Lipids and Lipoproteins, and Risk of Ischemic Heart Disease A Meta-Analysis Hans H. Wittrup, MD, PhD; Anne Tybjærg-Hansen, MD, DMSc; Børge G. Nordestgaard, MD, DMSc Background We assessed in meta-analyses the effect of the Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions in lipoprotein lipase in the heterozygous state on lipid metabolism and risk of ischemic heart disease (same order used below). Methods and Results In 29 separate studies, white subjects were screened for 1 of these substitutions; each meta-analysis included only some of these individuals. In population-based studies, heterozygote frequencies ranged from 0.04% to 0.2%, 2% to 4%, 1% to 7%, and 17% to 22% for the respective substitutions. Postheparin plasma lipoprotein lipase activity decreased 53% (95% CI, 31% to 75%) (only 1 study), 30% (22% to 37%), and 22% (8% to 35%) and was unchanged at 4% ( 10% to 19%), respectively. Plasma triglycerides increased 78% (95% CI, 64% to 92%), 20% (9% to 33%), and 31% (20% to 43%) and decreased 8% (4% to 11%), respectively. HDL cholesterol decreased 0.25 mmol/l (0.18 to 0.32), 0.08 mmol/l (0.04 to 0.12), and 0.12 mmol/l (0.10 to 0.15) and increased 0.04 mmol/l (0.02 to 0.06), respectively. Odds ratios for ischemic heart disease were 4.9 (95% CI, 1.2 to 20) (only 1 study), 1.4 (0.8 to 2.4), 1.2 (0.9 to 1.5), and 0.8 (0.7 to 1.0), respectively. Subgroup analysis indicated that women with the Asn291Ser substitution may have an increased risk of ischemic heart disease. Conclusions These meta-analyses suggest that compared with noncarriers, carriers of the Gly188Glu, Asp9Asn, and Asn291Ser substitutions have an atherogenic lipoprotein profile, whereas carriers of the Ser447Ter substitution have a protective lipoprotein profile. Accordingly, risk of ischemic heart disease in heterozygous carriers is increased for Gly188Glu carriers; at most, the increase is borderline for Asp9Asn and Asn291Ser carriers; and risk is possibly decreased for Ser447Ter carriers. (Circulation. 1999;99: ) Key Words: lipoproteins genetics cholesterol heart diseases meta-analysis enzymes In the homozygous form, amino acid changing mutations in lipoprotein lipase may lead to the chylomicronemia syndrome, 1 characterized by severe hypertriglyceridemia and extremely low HDL levels and possibly by ischemic heart disease. 2 Heterozygosity for amino acid changing mutations in this enzyme may therefore lead to intermediate levels of triglycerides and HDL and consequently to increased risk of ischemic heart disease. We explored this possibility. Most amino acid changing mutations in lipoprotein lipase are rare, either restricted to single families or isolated geographic regions 1 ; however, those leading to the Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions have been identified widely. 3 5 The Gly188Glu, Asp9Asn, and Asn291Ser substitutions are all located in the N-terminal end and may influence the catalytic activity of lipoprotein lipase, whereas the location of the Ser447Ter substitution in the C-terminal end may influence the enzyme-mediated uptake of lipoproteins by receptors on the cell surface. 3 The Ser447Ter substitution could therefore have effects quite different from those of the 3 other substitutions. In the homozygous form, the Gly188Glu substitution has been associated with chylomicronemia syndrome, 1,6,7 whereas homozygosity for the other 3 substitutions has at most a moderate effect on lipids and lipoproteins We assessed in meta-analyses 15 the effects of these 4 amino acid changing substitutions in the heterozygous state on lipoprotein lipase activity, plasma lipids and lipoproteins, and risk of ischemic heart disease. Mutations that do not change amino acids in lipoprotein lipase were not considered. When possible, subgroup analyses were performed for population-based studies as well as for men separately; only a few studies reported on women separately. These meta-analyses should be viewed as hypothesis generating and not hypothesis testing. Methods Study Selection The selection criteria for studies to be considered for these metaanalyses were as follows: (1) the Gly188Glu, Asp9Asn, Asn291Ser, Received September 22, 1998; revision received March 15, 1999; accepted March 26, From the Department of Clinical Biochemistry, Herlev University Hospital (H.H.W., A.T.-H.), Herlev, Denmark, and Department of Clinical Biochemistry, Glostrup University Hospital (B.G.N.), Glostrup, Denmark. Correspondence to Hans H. Wittrup, MD, PhD, Department of Clinical Biochemistry, Herlev University Hospital, DK-2730 Herlev, Denmark. hans@wittrup.net 1999 American Heart Association, Inc. Circulation is available at

2 2902 LPL Mutations and Ischemic Heart Disease Meta-Analyses of Effects of Lipoprotein Lipase Substitutions Substitution in Lipoprotein Lipase Gly188Glu Postheparin P-Lipoprotein Lipase Activity, % P-Cholesterol, mmol/l P-Apolipoprotein B, mg/dl n (carriers/noncarriers) 9/3 78/ / / / Carriers vs noncarriers 53 ( 75 to 31)* 0.1 ( 0.2 to 0.3) 0.0 ( 0.4 to 0.4)* 8 (0 to 15) 5 ( 7 to 17)* Asp9Asn n (carriers/noncarriers) 11/281 11/ / / / /4782 Carriers vs noncarriers 30 ( 37 to 22) 30 ( 37 to 22) 0.0 ( 0.1 to 0.1) 0.0 ( 0.1 to 0.1) 0 ( 4 to4) 0( 4to4) Asn291Ser n (carriers/noncarriers) 35/50 10/17 738/ / / / Carriers vs noncarriers 22 ( 36 to 7) 37 ( 61 to 13)* 0.0 ( 0.2 to 0.1) 0.1 ( 0.2 to 0.1) 2 ( 2 to6) 1( 2to4) Ser447Ter n (carriers/noncarriers) 138/ / / / / /2186 Carriers vs noncarriers 4 ( 10 to 19) 4 ( 10 to 19) 0.1 ( 0.2 to 0.0) 0.1 ( 0.2 to 0.0) 4 ( 6 to 1) 4 ( 6to 1) P indicates plasma. Values are shown as weighted mean difference (95% CI). *Based on 1 study only. or Ser447Ter substitutions in lipoprotein lipase were studied; (2) participants were white; (3) family, cross-sectional, case-control, or case-referent studies were included that identified both heterozygous carriers and noncarriers; (4) data were reported on at least 1 of the following variables: plasma levels of postheparin lipoprotein lipase activity, triglycerides, HDL cholesterol, apolipoprotein AI, total cholesterol, or apolipoprotein B or on the risk of ischemic heart disease; and (5) only data published as part of full articles in peer-reviewed journals were considered. Search Strategy Literature searches through January 1998 included Medline search, Embase search, reference lists of articles already on file, and the Cochrane Collaboration Library database. A total of 29 studies including whites were included, 6 14,16 34 whereas 25 studies were excluded (tables available from authors). Data Collection Data were collected as they appeared in the original studies; however, in a few studies, 6,11,19,24,26,27 we had to compute means with SDs or ORs with 95% CIs (Prism version 2.0, Graph-Pad Software Inc). For some studies of the Ser447Ter substitution, 10,21,34 it was not possible to separate a small number of homozygous from heterozygous individuals, and therefore these homozygous individuals were included in the present analyses. Analysis of Results Meta-analyses were performed on Review Manager version 3.0 (the Cochrane Collaboration; ftp://ftp.cochrane.co.uk/pub/handbook). The measurement of lipoprotein lipase activity is not yet standardized internationally. However, the relative differences between carriers and noncarriers are comparable from study to study. Furthermore, meta-analysis for lipoprotein lipase activity did not show evidence of heterogeneity, indicating that the individual studies were comparable. To concentrate on differences between carriers and noncarriers, the meta-analysis of lipoprotein lipase activity was performed for carriers as a percent of that for noncarriers. We transformed triglycerides logarithmically 14 to approximately fit a normal distribution, an assumption made in the meta-analysis. A post hoc antilogarithm presented triglyceride results as the relative difference (ratio) between carriers and noncarriers. Meta-analyses using random-effects models were employed. This takes into consideration the within-study comparison between carriers and noncarriers, as well as differences between studies. The nonindependence between family members was not considered. The weighted mean difference was calculated based on weighting of individual results by the inverse variance 15,35 ; this limits the impact of studies with wide CIs. Likewise, ORs were weighted by the inverse variance; Mantel-Haenszel estimates were calculated across individual studies. 15,36,37 An approximated normality test (z test) was used to examine the aggregated results. 15 We did overall analyses of both sexes combined and subgroup analyses based on population-based studies alone. Only 2 studies reported separate data for women. 14,16 Thus, it was only possible to do subgroup analyses for men alone. Results Twenty-nine family, cross-sectional, case-control, or casereferent studies including whites were selected for these meta-analyses. Some studies reported on 1 substitution. Each meta-analysis included only some of the individuals. In population-based studies, heterozygote frequencies of the Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions among control individuals ranged from 0.04% to 0.2%, 2% to 4%, 1% to 7%, and 17% to 22%, respectively. Mean plasma levels of postheparin lipoprotein lipase activity, triglycerides, HDL cholesterol, apolipoprotein AI, total cholesterol, and apolipoprotein B in groups of noncarriers in these studies ranged from 50 to 289 mu/ml, 0.80 to 8.46 mmol/l, 0.93 to 1.73 mmol/l, 100 to 173 mg/dl, 4.4 to 8.9 mmol/l, and 72 to 142 mg/dl, respectively. The inferences drawn concerning the Gly188Glu substitution were based on a limited number of studies, most of which were studies of families. The estimated effects on postheparin plasma lipoprotein lipase activity and risk of ischemic heart disease among these carriers were each based on only 1 study and were therefore not meta-analyses. Postheparin Plasma Lipoprotein Lipase Activity The Gly188Glu substitution in the heterozygous state decreased postheparin plasma lipoprotein lipase activity by 53% (1 study only), approximately twice the reduction observed for the Asp9Asn and Asn291Ser substitutions, respectively

3 Wittrup et al June 8, Continued P-Apolipoprotein AI, mg/dl 54/ / ( 34 to 6) 24 ( 37 to 12)* 142/ / ( 13 to 2) 6 ( 11 to 2) 631/ / ( 9 to 4) 6 ( 8 to 4) 525/ /2186 2(0to4) 2(0to4) (Table). For the Ser447Ter substitution, lipoprotein lipase activity was statistically unchanged at 4%. Plasma Triglycerides Gly188Glu, Asp9Asn, and Asn291Ser heterozygous carriers had an average increase in plasma triglycerides of 78%, 20%, and 31%, respectively, whereas carriers of the Ser447Ter substitution had a decrease of 8% compared with noncarriers (Figure 1). In analyses including population-based studies only, the results were similar (Figure 2). This was also the case in the subanalysis of men only (data not shown), except that there was a more discrete effect of the Asn291Ser substitution among men, in whom it increased triglycerides by just 14%. Plasma HDL Cholesterol and Apolipoprotein AI Gly188Glu, Asp9Asn, and Asn291Ser heterozygous carriers had an average decrease in HDL cholesterol of 0.25, 0.08, and 0.12 mmol/l, respectively, whereas Ser447Ter carriers had an increase of 0.04 mmol/l compared with noncarriers (Figure 3). In analyses restricted to population-based studies, the effects were similar (Figure 4). This also applied to subanalysis on men only (data not shown), except that the effect of the Asn291Ser substitution in men was slightly less than in the overall analysis, with HDL cholesterol decreased by 0.09 mmol/l (0.06 to 0.13 mmol/l). Apolipoprotein AI levels among heterozygous carriers showed a similar pattern as that observed for HDL cholesterol (Table). Plasma Cholesterol and Apolipoprotein B There was no effect of the Gly188Glu, Asp9Asn, and Asn291Ser substitutions on plasma cholesterol, but surprisingly, the Ser447Ter substitution appeared to slightly decrease plasma cholesterol (0.1 mmol/l; Table), an effect also found for men alone (data not shown). Aggregated data showed no effect of the Asp9Asn and Asn291Ser substitutions on plasma apolipoprotein B; however, the Gly188Glu substitution had a borderline elevating Figure 1. Meta-analyses of all studies of effect of Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions on plasma triglycerides. Results are shown as percent change among carriers compared with noncarriers. Size of squares indicating mean values is proportional to weight that each study contributed to aggregated result. Aggregated means with 95% CIs are shown by diamond symbols. effect of 8 mg/dl (Table). The Ser447Ter substitution decreased apolipoprotein B levels by 4 mg/dl; this effect was also observed in the analysis of men alone (data not shown). Risk of Ischemic Heart Disease ORs for ischemic heart disease in heterozygous carriers of the Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions were 4.9 (1.2 to 19.6) (1 study only), 1.4 (0.8 to 2.4), 1.2 (0.9 to 1.5), and 0.8 (0.7 to 1.0), respectively (Figure 5). In analysis of men only, ORs for Asp9Asn, Asn291Ser, and Ser447Ter carriers were similar to those in the overall analysis (data not shown). The single study examining only women found a highly significant increase in risk of ischemic heart disease among carriers of the Asn291Ser substitution. 14

4 2904 LPL Mutations and Ischemic Heart Disease Figure 2. Meta-analyses of population-based studies of effect of Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions on plasma triglycerides. Results are depicted as in Figure 1. Results for Gly188Glu are based on 1 study only. Discussion The present meta-analyses suggest that risk of ischemic heart disease is increased in Gly188Glu carriers, that the increase is borderline at most for Asp9Asn and Asn291Ser carriers, and that risk is possibly decreased for Ser447Ter carriers. The result for the Gly188Glu substitution is based on 1 study only and is therefore not based on a meta-analysis. The rare Gly188Glu substitution (carrier frequency 0.06%) 18 has a relatively large effect on decreasing enzyme activity (1 study only), increasing plasma triglycerides, and decreasing plasma HDL cholesterol. Furthermore, the more common Asp9Asn and Asn291Ser substitutions (carrier frequencies 3% and 5%, respectively) have more moderate effects on these variables, whereas the very common Ser447Ter substitution (carrier frequency 20%) appears to have the smallest and opposite effect on these variables. Whether the Asn291Ser substitution doubles the risk of ischemic heart disease in women remains to be confirmed, because thus far only 1 study has demonstrated this effect in women. 14 In support of this effect occurring specifically in women, the aggregated effect of the Asn291Ser substitution on plasma triglycerides, apolipoprotein AI, and HDL cholesterol was smaller in the analyses of men alone than in the Figure 3. Meta-analyses of all studies of effect of Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions on plasma HDL cholesterol. Results are depicted as in Figure 1. overall analysis, which suggests that this substitution has a more pronounced effect in women than in men. Methodological Quality of Individual Although all of the individual studies selected for these meta-analyses were of high quality and were published in peer-reviewed journals, several problems related to the individual studies may have influenced the summarized results. (1) Some studies measured plasma lipids and lipoproteins in the nonfasting state 11,14 ; however, an association between a more severe phenotype in carriers of any of these amino acid substitutions and the postprandial state has never been established. (2) Interaction with other potentially important variables such as age, diabetes mellitus, alcohol consumption, smoking habits, body mass index, digestion of saturated versus unsaturated fat, exercise, certain medication, and for women, menopausal status and use of hormonal replacement therapy may be important for the results regarding lipids and lipoproteins. Some of the studies included in the present meta-analyses presented plasma lipid and lipoprotein values

5 Wittrup et al June 8, Figure 5. Meta-analyses of population-based studies of association between Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions and risk of ischemic heart disease (IHD). Size of squares indicating mean values is proportional to weight that each study contributed to aggregated result. Aggregated ORs with 95% CIs are shown by diamond symbols. Results for Gly188Glu are based on 1 study only. Figure 4. Meta-analyses of population-based studies of effect of Gly188Glu, Asp9Asn, Asn291Ser, and Ser447Ter substitutions on plasma HDL cholesterol. Results are depicted as in Figure 1. Results for Gly188Glu are based on 1 study only. adjusted for some or all of these variables; however, although the consequence of such adjustment may have been to remove the effect of the amino acid substitution observed for unadjusted values, aggregated data nevertheless suggested that there was indeed a detectable effect of these 4 amino acid substitutions. (3) Ischemic heart disease status was not reported with identical diagnostic criteria but either specifically as coronary artery disease (indicating that angiography was performed) or nonfatal myocardial infarction, or more broadly as ischemic heart disease. This difference in reporting disease status, however, is unlikely to present a major problem because in every study, carriers and noncarriers were diagnosed by use of identical criteria. (4) It was not possible to exclude homozygous individuals from all the reported data 10,12,21 ; however, although these homozygous carriers may have experienced greater effects on triglycerides and HDL cholesterol than heterozygous carriers, it is likely that the effect of this small number of individuals was diluted among the many heterozygous individuals. (5) We included both family studies and population-based studies. It therefore can be questioned how these 2 types of samples can be combined in a meaningful analysis and for what population of inference the results are relevant, if any. However, we found only minimal evidence of heterogeneity in analyses in which the 2 types of studies were combined. 15 Furthermore, in subanalysis of population-based studies only, the results were similar to those in which the 2 types of studies were combined. (6) Finally, effects of these substitutions may be context dependent and thus vary from study to study. Biological Mechanism Lipoprotein lipase is believed to be organized in an N-domain (residues 1 to 312), which is important for the catalytic function of the enzyme, and a C-domain (residues 313 to 448), which is important for the lipoprotein lipase mediated uptake of lipoproteins by receptors on the cell surface. 3 This may help explain the effects of the 4 amino acid substitutions. The Gly188Glu substitution is located in the lipid-binding region: reduced binding and thus degradation of triglycerides may affect plasma levels of triglycerides and HDL cholesterol. The Asp9Asn substitution at the N-terminal end is not in a region with a known function but is situated near a glycosylation site that may influence overall catalytic activity. The Asn291Ser substitution is located in a heparinbinding cluster and may thus affect the interaction of lipoprotein lipase with the cell wall glycosaminoglycans. These 3 amino acid substitutions are located in the N-domain and therefore may reduce enzyme activity and consequently increase triglyceride levels, whereas the Ser447Ter substitution is located in the C-domain and thus may cause increased binding affinity of the shortened lipoprotein lipase to receptors or may affect its subunit interaction, either facilitating or otherwise affecting the formation of dimers, which would explain the opposite effect of this substitution compared with the other 3. Mechanistically, it seems plausible that lipoprotein lipase substitutions leading to decreased enzyme activity and consequently elevated triglyceride levels and reduced HDL cholesterol levels should increase an individual s risk of ischemic heart disease. Both increased plasma triglycerides and reduced HDL cholesterol levels are well-known cardiovascular risk fac-

6 2906 LPL Mutations and Ischemic Heart Disease tors. 38,39 Elevated triglycerides indicate that IDLs, VLDLs, and/or chylomicron remnants are present in plasma, and these particles may be selectively retained in the intima 40 and consequently promote atherosclerosis. The exact mechanism by which a mutation in lipoprotein lipase should result in remnant accumulation, however, is not clear. Reduced HDL cholesterol may result in reduced reverse cholesterol transport, indirectly promoting atherosclerosis. 39 Mutations in lipoprotein lipase resulting in increased triglyceride levels could even promote atherosclerosis by still other mechanisms (for example, by their association with small, dense LDL particles; by particularly promoting postprandial triglyceride-rich lipoproteins; or via an influence on hemostasis 38 ). Potential Confounding Linkage disequilibrium with other causative mutations nearby cannot be excluded completely. For the Gly188Glu and Asn291Ser substitutions, however, none have ever been identified. In contrast, the HindIII polymorphism in intron 8 of the lipoprotein lipase gene (strongly associated with altered lipid levels 41 ) seems to be in almost complete linkage disequilibrium with the Ser447Ter substitution 4,33 ; however, the Ser447Ter substitution is believed to be more important for the observed effects simply because it truncates 2 amino acids, whereas the HindIII polymorphism is an intron variant. Nevertheless, it is also possible that the HindIII polymorphism is in linkage disequilibrium with another hithertoundescribed variant in lipoprotein lipase or another nearby locus. 12 Recently, a promoter mutation (T-93G) was described in a patient with familial combined hyperlipidemia and the Asp9Asn substitution, 42 but at this point no evidence supports a causative role of the promoter mutation. 43 Future Research Efforts Although there is substantial evidence pointing toward these amino acid substitutions in lipoprotein lipase as important susceptibility mutations for small to moderate changes in plasma triglycerides, HDL cholesterol, and apolipoprotein AI, the evidence for the Gly188Glu substitution is only supported by data from 75 carriers and that for the Asp9Asn by data from 200 carriers, both considerably less than the number of carriers included in the comparisons for the Asn291Ser and Ser447Ter substitutions. Therefore, more population-based research to support the role of the Gly188Glu and Asp9Asn substitutions is required. Furthermore, the evidence for the effect of these substitutions on postheparin plasma levels of lipoprotein lipase mass and activity is very limited, and the effects on plasma cholesterol and apolipoprotein B need to be explored, in particular to strengthen the evidence for a possible explanation of a subset of patients with familial combined hyperlipidemia. 1 Although some evidence exists to support a direct association with ischemic heart disease, it is not strong, and within the area of a related disease, namely, ischemic cerebrovascular disease, it is almost absent. In addition, the potential differences related to sex and different ethnic groups, whether the effects of the substitutions are postprandial, and interactions with other common polymorphisms, eg, apolipoprotein E, need to be explored further. Acknowledgments This study was supported by the Danish Heart Foundation (grant No ), the Danish Medical Research Council (grant No ), and Chief Physician Johan Boserup and Lise Boserup s Fund. References 1. Brunzell J. Familial lipoprotein lipase deficiency and other causes of the chylomicronemia syndrome. In: Scriver C, Beaudet A, Sly W, Valle D, eds. The Metabolic and Molecular Bases of Inherited Disease. 7th ed. New York, NY: McGraw-Hill Inc; 1995: Benlian P, Gennes JLD, Foubert L, Zhang H, Gagne SE, Hayden MR. Premature atherosclerosis in patients with familial chylomicronemia caused by mutations in the lipoprotein lipase gene. N Engl J Med. 1996;335: Murthy V, Julien P, Gagne C. Molecular pathobiology of the human lipoprotein lipase gene. Pharmacol Ther. 1996;70: Hokanson JE. Lipoprotein lipase gene variants and risk of coronary disease: a quantitative analysis of population-based studies. Int J Clin Lab Res. 1997;27: Fisher RM, Humphries SE, Talmud PJ. Common variation in the lipoprotein lipase gene: effects on plasma lipids and risk of atherosclerosis. Atherosclerosis. 1997;135: Paulweber B, Wiebusch H, Miesenboeck G, Funke H, Assmann G, Hoelzl B, Sippl MJ, Friedl W, Patsch JR, Sandhofer F. Molecular basis of lipoprotein lipase deficiency in two Austrian families with type I hyperlipoproteinemia. Atherosclerosis. 1991;86: Wilson DE, Emi M, Iverius P-H, Hata A, Wu LL, Hillas E, Williams RR, Lalouel JM. Phenotypic expression of heterozygous lipoprotein lipase deficiency in the extended pedigree of a proband homozygous for a missense mutation. J Clin Invest. 1990;86: Galton DJ, Mattu R, Needham EW, Cavanna J. Identification of putative beneficial mutations for lipid transport. Z Gastroenterol. 1996;34(suppl 3): Gerdes C, Fisher RM, Nicaud V, Boer J, Humphries SE, Talmud PJ, Faergeman O. Lipoprotein lipase variants D9N and N291S are associated with increased plasma triglyceride and lower high-density lipoprotein cholesterol concentrations: studies in the fasting and postprandial states: the European Atherosclerosis Research. Circulation. 1997;96: Groenemeijer BE, Hallman MD, Reymer PW, Gagne E, Kuivenhoven JA, Bruin T, Jansen H, Lie KI, Bruschke AV, Boerwinkle E, Hayden MR, Kastelein JJ. Genetic variant showing a positive interaction with betablocking agents with a beneficial influence on lipoprotein lipase activity, HDL cholesterol, and triglyceride levels in coronary artery disease patients: the Ser447-stop substitution in the lipoprotein lipase gene: REGRESS Study Group. Circulation. 1997;95: Mailly F, Tugrul Y, Reymer PW, Bruin T, Seed M, Groenemeyer BF, Asplund-Carlson A, Vallance D, Winder AF, Miller GJ, Kastelein JJP, Hamsten A, Olivecrona G, Humphries SE, Talmud PJ. A common variant in the gene for lipoprotein lipase (Asp93Asn): functional implications and prevalence in normal and hyperlipidemic subjects. Arterioscler Thromb Vasc Biol. 1995;15: Mattu RK, Needham EW, Morgan R, Rees A, Hackshaw AK, Stocks J, Elwood PC, Galton DJ. DNA variants at the LPL gene locus associate with angiographically defined severity of atherosclerosis and serum lipoprotein levels in a Welsh population. Arterioscler Thromb. 1994;14: Salah D, Bohnet K, Gueguen R, Siest G, Visvikis S. Combined effects of lipoprotein lipase and apolipoprotein E polymorphisms on lipid and lipoprotein levels in the Stanislas cohort. J Lipid Res. 1997;38: Wittrup HH, Tybjaerg-Hansen A, Abildgaard S, Steffensen R, Schnohr P, Nordestgaard BG. A common substitution (Asn291Ser) in lipoprotein lipase is associated with increased risk of ischemic heart disease. J Clin Invest. 1997;99: Mulrow CD, Oxman AD, eds. Cochrane Collaboration Handbook. Oxford, UK: Cochrane Collaboration; 1996: Pimstone SN, Gagne SE, Gagne C, Lupien PJ, Gaudet D, Williams RR, Kotze M, Reymer PW, Defesche JC, Kastelein JJ, Moorjani S, Hayden MR. Mutations in the gene for lipoprotein lipase: a cause for low HDL cholesterol levels in individuals heterozygous for familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 1995;15: Sprecher DL, Harris BV, Stein EA, Bellet PS, Keilson LM, Simbartl LA. Higher triglycerides, lower high-density lipoprotein cholesterol, and

7 Wittrup et al June 8, higher systolic blood pressure in lipoprotein lipase-deficient heterozygotes: a preliminary report. Circulation. 1996;94: Nordestgaard BG, Abildgaard S, Wittrup HH, Steffensen R, Jensen G, Tybjærg-Hansen A. Heterozygous lipoprotein lipase deficiency: frequency in the general population, effect on plasma lipid levels, and risk of ischemic heart disease. Circulation. 1997;96: Elbein SC, Yeager C, Kwong LK, Lingam A, Inoue I, Lalouel JM, Wilson DE. Molecular screening of the lipoprotein lipase gene in hypertriglyceridemic members of familial noninsulin-dependent diabetes mellitus families. J Clin Endocrinol Metab. 1994;79: de Bruin TW, Mailly F, van Barlingen HH, Fisher R, Castro Cabezas M, Talmud P, Dallinga-Thie GM, Humphries SE. Lipoprotein lipase gene mutations D9N and N291S in four pedigrees with familial combined hyperlipidaemia. Eur J Clin Invest. 1996;26: Zhang Q, Cavanna J, Winkelman BR, Shine B, Gross W, Marz W, Galton DJ. Common genetic variants of lipoprotein lipase that relate to lipid transport in patients with premature coronary artery disease. Clin Genet. 1995;48: Jukema JW, van Boven AJ, Groenemeijer B, Zwinderman AH, Reiber JHC, Bruschke AVG, Henneman JA, Molhoek GP, Bruin T, Jansen H, Gagné E, Hayden MR, Kastelein JJP. The Asp9Asn mutation in the lipoprotein lipase gene is associated with increased progression of coronary atherosclerosis. Circulation. 1996;94: Mailly F, Fisher RM, Nicaud V, Luong L-A, Evans AE, Marques-Vidal P, Luc G, Arveiler D, Bard JM, Poirier O, Talmud PJ, Humphries SE. Association between the LPL-D9N mutation in the lipoprotein lipase gene and plasma lipid traits in myocardial infarction survivors from the ECTIM study. Atherosclerosis. 1996;122: Zhang H, Reymer PW, Liu MS, Forsythe IJ, Groenemeyer BE, Frohlich J, Brunzell JD, Kastelein JJ, Hayden MR, Ma Y. Patients with apoe3 deficiency (E2/2, E3/2, and E4/2) who manifest with hyperlipidemia have increased frequency of an Asn 2913Ser mutation in the human LPL gene. Arterioscler Thromb Vasc Biol. 1995;15: Reymer PW, Gagne E, Groenemeyer BE, Zhang H, Forsyth I, Jansen H, Seidell JC, Kromhout D, Lie KE, Kastelein J, Hayden MR. A lipoprotein lipase mutation (Asn291Ser) is associated with reduced HDL cholesterol levels in premature atherosclerosis. Nat Genet. 1995;10: Fisher RM, Mailly F, Peacock RE, Hamsten A, Seed M, Yudkin JS, Beisiegel U, Feussner G, Miller G, Humphries SE, Talmud PJ. Interaction of the lipoprotein lipase asparagine 2913serine mutation with body mass index determines elevated plasma triacylglycerol concentrations: a study in hyperlipidemic subjects, myocardial infarction survivors, and healthy adults. J Lipid Res. 1995;36: Jemaa R, Fumeron F, Poirier O, Lecerf L, Evans A, Arveiler D, Luc G, Cambou JP, Bard JM, Fruchart JC, Apfelbaum, Cambien F, Tiret L. Lipoprotein lipase gene polymorphisms: associations with myocardial infarction and lipoprotein levels: the ECTIM study. J Lipid Res. 1995; 36: Reymer PW, Groenemeyer BE, Gagne E, Miao L, Appelman EE, Seidel JC, Kromhout D, Bijvoet SM, van de Oever K, Bruin T, Hayden MR, Kastelein JJP. A frequently occurring mutation in the lipoprotein lipase gene (Asn291Ser) contributes to the expression of familial combined hyperlipidemia. Hum Mol Genet. 1995;4: Syvänne M, Antikainen M, Ehnholm S, Tenkanen H, Lahdenperä S, Ehnholm C, Taskinen M-R. Heterozygosity for Asn 291 -Ser mutation in the lipoprotein lipase gene in two Finnish pedigrees: effect of hyperinsulinemia on the expression of hypertriglyceridemia. J Lipid Res. 1996;37: Hoffer MJV, Bredie SJH, Boomsma DI, Reymer PWA, Kastelein JJP, de Knijff P, Demacker PNM, Stalenhoef AFH, Havekes LM, Frants RR. The lipoprotein lipase (Asn291-Ser) mutation is associated with elevated lipid levels in families with familial combined hyperlipidemia. Atherosclerosis. 1996;119: Pimstone SN, Clee SM, Gagne SE, Miao L, Zhang H, Stein EA, Hayden MR. A frequently occurring mutation in the lipoprotein lipase gene (Asn291Ser) results in altered postprandial chylomicron triglyceride and retinyl palmitate response in normolipidemic carriers. J Lipid Res. 1996; 37: Knudsen P, Murtomäki S, Antikainen M, Ehnholm S, Lahdenperä S, Ehnholm C, Taskinen M-R. The Asn-2913Ser and Ser-4473Stop mutations of the lipoprotein lipase gene and their significance for lipid metabolism in patients with hypertriglyceridaemia. Eur J Clin Invest. 1997;27: Stocks J, Thorn JA, Galton DJ. Lipoprotein lipase genotypes for a common premature termination codon mutation detected by PCRmediated site-directed mutagenesis and restriction digestion. J Lipid Res. 1992;33: Peacock RE, Hamsten A, Nilsson-Ehle P, Humphries SE. Associations between lipoprotein lipase gene polymorphisms and plasma correlations of lipids, lipoproteins and lipase activities in young myocardial infarction survivors and age-matched healthy individuals from Sweden. Atherosclerosis. 1992;97: Laird NM, Mosteller F. Some statistical methods for combining experimental results. Int J Technol Assess Health Care. 1990;6: Armitage P, Berry G, eds. Statistical Methods in Medical Research. Oxford, UK: Blackwell Science Ltd; 1994:66 71, 85 88, , , Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Institute. 1959;22: Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies. J Cardiovasc Risk. 1996;3: Barter PJ, Rye K-A. High density lipoproteins and coronary heart disease. Atherosclerosis. 1996;121: Nordestgaard BG, Wootton R, Lewis B. Selective retention of VLDL, IDL, and LDL in the arterial intima of genetically hyperlipidemic rabbits in vivo: molecular size as a determinant of fractional loss from the intima-inner media. Arterioscler Thromb Vasc Biol. 1995;15: Galton DJ. Common genetic determinants of dyslipidemia: the hypertriglyceridemia/low-high-density lipoprotein syndrome. J Cardiovasc Pharmacol. 1995;25(suppl 4):S35 S Yang WS, Nevin DN, Peng R, Brunzell JD, Deeb SS. A mutation in the promoter of the lipoprotein lipase (LPL) gene in a patient with familial combined hyperlipidemia and low LPL activity. Proc Natl Acad Sci USA. 1995;92: Hall S, Chu G, Miller G, Cruickshank K, Cooper JA, Humphries SE, Talmud PJ. A common mutation in the lipoprotein lipase gene promoter, 93T/G, is associated with lower plasma triglyceride levels and increased promoter activity in vitro. Arterioscler Thromb Vasc Biol. 1997;17: