| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Genetic Epidemiology Unit (A.I., Y.S.A., C.M.v.D.), Departments of Epidemiology and Biostatistics (A.H., B.H.C.S., J.C.M.W.), Clinical Genetics (B.A.O.), and Metabolic and Vascular Diseases (E.J.G.S.), Erasmus Medical Center, 3000 CA Rotterdam, The Netherlands; Department of Pharmacovigilance (F.A.S.-T.), The Dutch Medicines Evaluation Board Agency, 2500 BE The Hague, The Netherlands; and Department of Pharmacoepidemiology and Pharmacotherapy (O.H.K., A.-H.M.v.d.Z.), Utrecht Institute of Pharmaceutical Sciences, Utrecht University, 3508 TB Utrecht, The Netherlands
Address all correspondence and requests for reprints to: Prof. C. M. van Duijn, Department of Epidemiology and Biostatistics, Erasmus Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. E-mail: c.vanduijn{at}erasmusmc.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Design: The study was conducted in the Rotterdam Study, a large single-center prospective cohort study in people aged 55 yr and older. Lipid levels were analyzed using linear regression models, and risk of MI was assessed with Cox proportional hazards models.
Results: The hepatic lipase variant was associated with an increase in serum HDL-c levels of 0.11 mmol/liter in both genders, whereas an increased risk of MI was observed only in men [hazard ratio, 1.32 (95% confidence interval, 1.05–1.66) for CT vs. CC and 1.75 (95% confidence interval, 1.39–2.20) for TT vs. CC]. This effect was independent of serum HDL-c. LIPC –514C > T interacted with CETP I405V with respect to serum HDL-c concentrations. Those homozygous for both mutations saw a marked elevation in HDL-c levels (0.29 mmol/liter, Pinteraction = 0.05). These increased HDL-c levels, however, were not inversely associated with atherosclerosis or MI risk.
Conclusions: LIPC genotype affects HDL-c levels and risk of MI in males. The interaction of this variant with CETP on HDL-c levels helps elucidate the underlying mechanisms and suggests that the beneficial effects of CETP inhibition may vary in particular subgroups.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The genes encoding hepatic lipase (LIPC) and cholesteryl ester transfer protein (CETP) are consistently associated with circulating HDL-c. The T allele of the –514 C > T single nucleotide polymorphism (SNP) in the promoter region of the LIPC gene is associated with a substantial decrease in hepatic lipase activity and a modest, but significant, elevation of HDL-c levels (4). The V allele of the CETP I405V polymorphism, which leads to an isoleucine to valine substitution in the 405th residue of the CETP protein, is strongly associated with decreased CETP activity and mass. Through this mechanism, the CETP SNP causes significant increases in HDL-c levels (5).
Both hepatic lipase and CETP are involved in modeling and remodeling HDL-c subspecies, particularly intermediate-sized 
-1 particles (2). CETP-mediated triglyceride enrichment of HDL-c notably increases the ability of hepatic lipase to model HDL-c (6). As such, the potential interaction between LIPC and CETP variants on lipid levels are of interest. In interaction studies published to date, the joint effects of the LIPC –514C > T genotype with the CETP Taq1B and –1337C > T polymorphisms revealed no significant evidence for epistasis (7, 8, 9).
Although the relationships between LIPC and CETP and HDL-c are well established, their roles in modulating cardiovascular disease risk are less clear. Previous findings on the effect of the LIPC –514 C > T polymorphism on MI risk were inconsistent. Two studies found no association (10, 11), although the latter study showed a borderline deleterious effect of the T allele. In two other reports, the T allele was variously associated with a protective effect on MI risk in coronary heart disease patients (12) and an increased risk of acute MI in males (13). This second finding is counterintuitive, because the T allele is associated with increased HDL-c levels.
Most previous studies examining the relationship between CETP variants and MI risk have focused on the intronic Taq1B variant. One of these found no effect on MI risk, (14), whereas others found effects in the anticipated direction (increased HDL-c and decreased MI risk with the CETP Taq1B B2 allele) (15, 16). The role of the CETP I405V variant is similarly unclear. Several studies evaluated the impact of the CETP I405V SNP on MI risk. Of these, some found no effect (17, 18), and another observed no differences in allele frequencies between children of early MI patients and controls (19). The CETP I405V polymorphism was previously studied in the Rotterdam Study and was associated with an 0.06 mmol/liter increase in HDL-c levels in both men and women and a significant decrease in MI risk in males (20).
Epistatic effects that predispose one to incidence of coronary disease, such as MI, are increasingly being explored (21). Previous studies suggested a possible interaction between CETP and LIPC variants with respect to coronary artery disease markers (angiography and coronary stenosis index) (8, 9). To date, no research addressing this potential interaction in terms of modifying MI risk has been published, and population-based data with respect to MI are also lacking. An interaction between these LIPC and CETP variants, which decrease protein activity and increase HDL-c levels, should, in theory, decrease atherosclerosis and reduce the incidence of MI in our prospective cohort study if the function of the resulting HDL-c particles remains unchanged.
The present study was conducted with two aims. The first was to assess the impact of the LIPC –514 C > T polymorphism with respect to serum lipid levels, atherosclerosis proxies, and incidence of MI. The second objective was to determine whether there was evidence for an interaction between this polymorphism and the CETP I405V polymorphism.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This study was embedded in the Rotterdam Study. Previous descriptions of this population have been published (22). Briefly, the Rotterdam Study is a single-center prospective cohort comprising 7983 individuals aged 55 yr and older at the inception of the study. Baseline examinations took place between 1990 and 1993. All participants completed written informed consents, and the Medical Ethics Committee at Erasmus University approved the protocols for the ascertainment and examination of human subjects. At their baseline examination, these individuals completed interviews with a trained research assistant, including details of alcohol consumption, smoking, hormone replacement therapy, lipid-lowering therapy, and diabetes status, and underwent medical examinations at the research center. Serum lipid measurements of total cholesterol (TC) and HDL-c were determined enzymatically, using an automated procedure (23). Common carotid intima media thickness (IMT) was assessed by ultrasonography, as described previously (24). Carotid plaque score was determined by the number of sites (common carotid, internal carotid, and bifurcation on both the left and right sides) that showed visible focal widening protruding into the luminal space and was scored from zero to six.
Follow-up data collection between baseline (1990–1993) and January 1, 2005 included data on the incidence of MI. Information on fatal and nonfatal MIs was obtained from general practitioners through a computerized reporting system. Two research physicians examined the patients’ medical records and verified the events. When these physicians disagreed, a medical expert in the field determined the diagnosis. In the case of multiple events, the first event was used for this analysis; prevalent MIs were excluded.
Genotyping
Genomic DNA was extracted from whole-blood samples drawn at the baseline examination, using the salting out method. Subjects (n = 6571) were genotyped for both polymorphisms with a TaqMan allelic discrimination Assay-By-Design (Applied Biosystems, Foster City, CA). Forward and reverse primer sequences for the I405V SNP were 5'-CTCACCATGGGCATTTGATTGG-3' and 5'-CGGTGATCATTGACTGCAGGAA-3', respectively. The minor groove binding probes were 5'-CTCCGAGTCCGTCCAGA-VIC-3' and 5'- TCCGAGTCCATCCAGA-FAM-3'. For the LIPC SNP, forward and reverse primers were 5'-TTTGCTTCTTCGTCAGCTCCTT-3' and 5'-GTCAAAGTGTGGTGCAGAAAACC-3'; probes were 5'-CTTCACCCCCATGTCAA-VIC-3' and 5'-TTCACCCCCGTGTCAA-FAM-3'.
The assays used 5 ng genomic DNA and 5 µl reaction volumes. The amplification and extension protocol was as follows: an initial activation step of 10 min at 95 C preceded 40 cycles of denaturation at 95 C for 15 sec and annealing and extension at 50 C for 60 sec. Allele-specific fluorescence was then analyzed on an ABI Prism 7900HT Sequence Detection System with SDS version 2.1 (Applied Biosystems).
Statistical analysis
Deviations from Hardy-Weinberg proportions were evaluated using an exact test (25). General characteristics were compared using univariate ANOVA for continuous variables and 
2 for dichotomous variables. After stratification by gender, serum lipid outcomes (TC, HDL-c, and TC/HDL-c ratio) and carotid IMT and plaques were examined with respect to the two polymorphisms. These analyses were conducted with a general linear model adjusting for age, smoking, alcohol consumption, body mass index (BMI), lipid-lowering therapy, and diabetes mellitus. Because the residuals for the lipid measures and IMT were not normally distributed, these values were log transformed to improve their normality.
Cox proportional hazards models were implemented to assess the relationship between the polymorphisms and incidence of MI, using follow-up time on the independent axis. The data were stratified by gender, and three models were fitted. The first was adjusted only for age. In the second model, the effect of the genotype was adjusted for classical MI risk factors, including age, smoking, BMI, systolic and diastolic blood pressure, diabetes mellitus, and total serum cholesterol. In the third model, adjustments were made for these classical risk factors and, additionally, serum HDL-c levels to assess effects of the gene independent of HDL-c.
To test for interaction between the genes, all possible combinations of genotypes of the two SNPs were analyzed. Models with/without the combined effect were then tested for differences using a likelihood ratio test.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As published previously, baseline characteristics did not differ between genotypes with respect to the CETP I405V polymorphism (20). Similarly, no differences for these traits were observed by LIPC genotype, with the exception of systolic blood pressure. Systolic blood pressure was highest in individuals heterozygous for the T allele and lowest in TT carriers (Table 1
).
|
). The difference between homozygous groups was 0.11 mmol/liter in both genders (P < 0.001), whereas heterozygotes saw more modest increases (0.04 mmol/liter in males and 0.03 mmol/liter in females). Subsequent adjustment for factors known to influence serum HDL-c levels did not change these findings. No statistically significant differences were noted for either genotype for TC. As a consequence of the association with serum HDL-c, the serum TC/HDL-c ratio was decreased in LIPC-T carriers in both genders (Pmale < 0.05 and Pfemale = 0.01). Despite these changes in HDL-c and TC/HDL-c ratio, no differences were observed for the common carotid IMT and plaque score (Table 2). As described previously, the CETP I405V SNP was associated with an increase of 0.06 mmol/liter in serum HDL-c in both genders and a decrease in serum TC/HDL-c ratio of 0.18 in males and 0.24 in females (20).
|
). In females, no association was found. In males, the T allele of the LIPC genotype was associated with an increase in MI hazard (Ptrend = 0.02), despite the finding that this allele was associated with high serum HDL-c levels (Table 2
). The hazard for CT carriers was 1.32 [95% confidence interval (CI), 1.05–1.66] and for TT carriers was 1.75 (95% CI, 1.39–2.20). Inclusion of classical risk factors in the model did not alter these results [hazard ratio (HR) for CT of 1.30 (95% CI, 1.03–1.65) and HRTT of 1.70 (95% CI, 1.35–2.15)]. These hazards remained significantly increased after inclusion of HDL-c in the model [HRCT of 1.35 (95% CI, 1.07–1.70) and HRTT of 1.82 (95% CI, 1.44–2.30); Ptrend = 0.01] (Fig. 1). As described previously, the V allele of the CETP I405V polymorphism significantly decreased MI risk in males [HRIV of 0.75 (95% CI, 0.59–0.95) and HRVV of 0.56 (95% CI, 0.44–0.71)].
|
). The mean difference in serum HDL-c between LIPC-TT/CETP-VV carriers compared with the LIPC-CC/CETP-II genotype group was 0.29 mmol/liter. This value is considerably greater than expected from the sum of the mean differences in serum HDL-c attributable to the individual genes (CETP-VV vs. CETP-II, 0.02 mmol/liter; LIPC-TT vs. LIPC-CC, 0.06 mmol/liter; Pinteraction = 0.05). Intermediate combinations (i.e. heterozygous for one or both variants) also showed increases larger than expected for the two SNPs individually. These patterns were similar after stratification by gender (Fig. 2), although the observed combined effect of the genes was stronger in women than in men (0.34 mmol/liter, P = 0.03 and 0.21 mmol/liter, P = 0.13, respectively). No significant differences were observed for IMT (data not shown).
|
). Although none of the hazards were significantly increased in either the overall analysis or the analysis stratified by gender, the patterns in MI hazard diverged from the patterns predicted from the effect of the two genes on serum HDL-c (Fig. 2). The point estimate for MI risk was highest in those who carried the combination of the CETP-II and LIPC-TT genotypes, whereas this genotype combination was associated with higher serum HDL-c levels compared with the CETP-II and LIPC-CC combination. When studying the combined effects of CETP and LIPC, the MI hazard decreased with the number of CETP V alleles, as predicted by the effect of this SNP on serum HDL-c levels. In contrast, the hazard of MI increased with the number of LIPC T alleles, although none of these HRs were significantly increased, overall or in men. The only exception to these trends was in the CETP-VV/LIPC-TT group, which showed the lowest risk. This group is very small, however, making the risk estimate unreliable.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
When studying the interaction between the LIPC –514C > T polymorphism and the CETP I405V polymorphism, there was significant evidence for interaction between the two SNPs, leading to marked increases in serum HDL-c in those individuals homozygous for the minor allele of both variants. This large departure from additivity was statistically significant and suggests epistasis between the two genes. A plausible mechanism for this interaction may result from the decreased catabolic efficacy of hepatic lipase with respect to triglyceride-poor HDL-c (3). Decreases in CETP activity attributable to I405V genotype would lead to diminished CETP-mediated triglyceride enrichment of HDL-c and a subsequent reduction in hepatic lipase efficiency, in addition to the decrease in lipase activity attributable solely to –514C > T genotype.
This interaction may have consequences in terms of therapeutic CETP inhibition, which is currently the subject of intense interest (26, 27). Because the LIPC variant seems to modify the effects of the CETP variant, it is likely that the LIPC genotype may also be involved in the response to pharmaceutical CETP inhibition. Recently, however, the clinical trial of a CETP inhibitor, torcetrapib, ceased because the drug caused excess mortality (28). Slight increases in systolic blood pressure were noted in early trials and may explain these adverse effects. Alternatively, HDL-c may become dysfunctional during CETP inhibition. At the least, these results suggest that this topic warrants additional study.
The interaction between CETP and LIPC, however, did not translate into the expected reduction in risk of MI. While studying the joint effect of the two genes, no significant associations with the hazard of MI could be observed in either the overall or gender-specific analysis. The lack of association in the interaction model may be explained in large part by the lack of statistical power to study interactions, (29) even for a large study, such as the Rotterdam Study. Specifically, the number of dual rare homozygotes, and the number of events observed in this subgroup, was insufficient to analyze this interaction in terms of MI hazard.
Several hypotheses may help to explain the seemingly paradoxical associations between the effects of LIPC on circulating HDL-c and MI. One potential explanation revolves around the fact that different HDL-c subspecies may result in different risk levels (30) and that some subfractions may be atherogenic under some conditions (31). Unfortunately, HDL-c subfractions were not determined in the Rotterdam Study. Another explanation for the apparently contradictory results is that the increase in MI risk by LIPC genotype is mediated by the nonlipolytic function of hepatic lipase, which may enhance atherosclerotic development. LIPC can reduce remnant and low-density lipoprotein cholesterol (LDL-c) through this noncatalytic pathway. This reduction, demonstrated in mouse models (32, 33), implies that benefits derived from increased serum HDL-c may be outweighed by decreased clearance of apolipoprotein B-containing particles. A third plausible explanation might relate to the effects of hepatic lipase in terms of the oxidation status of LDL-c particles, because oxidized LDL-c particles are thought to play an important role in atherogenesis (34, 35). A study of LIPC –514C > T in familial combined hyperlipidemia observed that the T allele led to increases in malondialdehyde-modified LDL-c (36), a species of oxidized LDL-c that may be a useful biomarker for atherosclerosis (37). Assuming that this process also occurs in nonaffected individuals might explain the results obtained in the current study.
In this study, increased HDL-c was observed in both genders, but increased MI risk was demonstrated only in men. Previous studies on CETP and LIPC genetic variation also noted gender differences (38, 39). Differences in sex steroid hormone levels and activity are one possible explanation, although these women were postmenopausal. Additionally, women tend to experience MI later in life than men (40) and are, therefore, less likely to become cases during the follow-up period; the noted differences might be attributable to this factor. Lastly, the possibility of a chance finding cannot be excluded.
The apparent discrepancy between the effect of LIPC genotype on HDL-c levels and risk of incidence of MI observed in men suggests that attempts to identify groups at high risk for MI based on genotypes associated with a favorable lipid profile may be hindered by unexpected associations and emphasizes the importance of studying clinically more relevant outcomes, such as MI. These efforts will be further complicated by the large number of, as yet, unidentified genes and interactions that are likely to affect both lipid levels and MI risk.
In summation, the T allele of the LIPC –514 C > T polymorphism was significantly associated with HDL-c levels in both genders and an increased risk of MI in males. The interaction between this SNP and the CETP I405V variant was also examined. Those individuals homozygous for both mutations (LIPC-TT/CETP-VV) possessed markedly increased serum HDL-c levels, but this did not appear to affect MI risk in the present study.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online April 17, 2007
Abbreviations: BMI, Body mass index; CETP, cholesteryl ester transfer protein; CI, confidence interval; HDL-c, high-density lipoprotein cholesterol; HR, hazard ratio; IMT, intima media thickness; LDL-c, low-density lipoprotein cholesterol; LIPC, polymorphism in the hepatic lipase gene; MI, myocardial infarction; SNP, single nucleotide polymorphism; TC, total cholesterol.
Received February 6, 2007.
Accepted April 9, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
T hepatic lipase promoter region polymorphism and plasma lipids: a metaanalysis. J Clin Endocrinol Metab 89:3858–3863This article has been cited by other articles:
 |
|
 |
|
  P Henneman, Y S Aulchenko, R R Frants, K W van Dijk, B A Oostra, and C M van Duijn Prevalence and heritability of the metabolic syndrome and its individual components in a Dutch isolate: the Erasmus Rucphen Family study J. Med. Genet., September 1, 2008; 45(9): 572 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Thompson, E. Di Angelantonio, N. Sarwar, S. Erqou, D. Saleheen, R. P. F. Dullaart, B. Keavney, Z. Ye, and J. Danesh Association of Cholesteryl Ester Transfer Protein Genotypes With CETP Mass and Activity, Lipid Levels, and Coronary Risk JAMA, June 18, 2008; 299(23): 2777 - 2788. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
