Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (54)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Klerkx, A. H.E.M.
Right arrow Articles by Kuivenhoven, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Klerkx, A. H.E.M.
Right arrow Articles by Kuivenhoven, J. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 2 111-123
© 2003 Oxford University Press

Haplotype analysis of the CETP gene: not TaqIB, but the closely linked -629C->A polymorphism and a novel promoter variant are independently associated with CETP concentration

Anke H.E.M. Klerkx1,*, Michael W.T. Tanck2,4, John J.P. Kastelein1, Henri O.F. Molhuizen1,5, J. Wouter Jukema3, Aeilko H. Zwinderman2,4 and Jan Albert Kuivenhoven1

1Department of Vascular Medicine and 4Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, Amsterdam, The Netherlands, 2Department of Medical Statistics and 3Department of Cardiology, Leiden University Medical Center (LUMC), Leiden, The Netherlands and 5Unilever Research Vlaardingen, Unilever Health Institute, Vlaardingen, The Netherlands

Received September 2, 2002; Accepted October 29, 2002


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The TaqIB polymorphism in intron 1 of the cholesteryl ester transfer protein (CETP) gene is associated with plasma CETP concentration, high-density lipoprotein cholesterol (HDL-C) and coronary artery disease (CAD). These associations are generally thought to arise from linkage disequilibrium between TaqIB and (an)other functional polymorphism(s). To identify putative functional sites, we investigated phenotypic associations of TaqIB and four tightly linked polymorphisms (novel -2708G->A and +784CCC->A, and previously identified -971G->A and -629C->A) in 709 males with CAD (REGRESS). In addition to genotype analyses, a novel method to estimate haplotype effects was used to examine the individual and joint effects of these DNA variants on CETP concentration and HDL-C. All polymorphisms were associated with CETP concentration and HDL-C, except for -971 with HDL-C. Stepwise regression and haplotype analyses indicated that only -629 was independently associated with HDL-C. Similar analyses additionally indicated that -2708 and -629 were independently associated with CETP concentration, whereby the most frequent alleles acted in a cumulative manner. Nonetheless, detailed haplotype analysis revealed that a 3-polymorphism haplotype model consisting of -2708, -629 and -971 explained the variation in CETP concentration best. The involvement of -971 could be due to interaction effects that were observed between -971 and both -629 (P<0.001) and -2708 (P=0.047). In conclusion, the TaqIB polymorphism is not instrumental in determining CETP or HDL-C levels, but is a marker for the -629 promoter variant. Our analyses, furthermore, indicate that the -2708 and -971 polymorphisms are likely to play a role in determining CETP concentration.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Cholesteryl ester transfer protein (CETP) plays a central role in human lipoprotein metabolism. This 74 000 kDa plasma protein shuttles cholesteryl esters (CE) from high-density lipoprotein (HDL) to Apolipoprotein B (ApoB) containing lipoproteins in exchange for triglycerides (1). Human CETP deficiency is characterized by very high HDL cholesterol (HDL-C) levels, but the role of CETP in determining HDL-C in the general population and how it may modulate the risk for coronary artery disease (CAD) is unclear (2–4). Through its function in the reverse cholesterol transport (RCT) pathway, CETP is potentially anti-atherogenic, because it may facilitate the removal of excess cholesterol from the body via LDL-receptor mediated uptake in the liver and excretion into the bile. On the other hand, as CETP may lower the concentration of atheroprotective HDL-C, it is often considered to be pro-atherogenic (4–7).

The CETP gene is highly polymorphic; many mutations in both coding and non-coding regions have been described (8–11). Of these, the TaqIB polymorphism in intron 1 has been studied extensively due to associations with HDL-C (12–16), but so far the significance of the effects of this polymorphism is poorly understood (17). Significant associations of the wild-type B1B1 genotype (TaqI restriction site present) with higher plasma CETP concentration and/or CETP activity and lower HDL-C were found in several large studies (11,14,16,18), but this is not consistently observed (12,19–21). Furthermore, it has been reported that the effects of TaqIB on the above parameters are gender dependent and also influenced by alcohol use, body mass index (BMI) and insulin levels (10–12,16,22,23). In addition to the associations of TaqIB with parameters in lipid metabolism, this polymorphism has been shown to be associated with the risk of CAD. In the Framingham Offspring Study, the B2 allele was associated with a reduced risk of coronary heart disease in men (16), and recently this was also shown in the Veterans Affair HDL-C Intervention Trial (VA-HIT) (24). Furthermore, it was shown that TaqIB predicted the response to lipid lowering therapy with respect to progression of coronary atherosclerosis, i.e. pravastatin was shown to be effective in retarding the progression only in men carrying the B1 allele (14).

The TaqIB polymorphism concerns a single nucleotide polymorphism (SNP) that affects the 277th nucleotide of intron 1 (902 bp). Due to its location, it is assumed that this SNP is not part of a functional regulatory site, but that it is a marker for another functional site or possibly multiple linked polymorphisms with independent effects (17). To date, several associations of the TaqIB site with other CETP variants have been investigated. A number of mutations resulting in amino acid substitutions, namely A373P, I405V and R451Q, were found to be independent variations from the TaqIB polymorphism with respect to effects on HDL-C and CETP activity (9,11,15,20,25). However, recently a promoter variant of the CETP gene was identified at position -629 relative to the transcription start (26). In the ECTIM study, the less common A allele was associated with lower plasma CETP concentration (P<0.0001) and independently with increased HDL (P<0.0001) (11,26). Moreover, a direct effect of variation at -629 on CETP gene expression was established in vitro using reporter gene expression in liver cells (26). Genotype analysis in several cohorts showed that the -629 polymorphism was almost fully concordant with TaqIB, indicating that this variation is a good candidate to explain the effects observed for the TaqIB polymorphism (11,20,26).

To determine whether the TaqIB polymorphism itself is putatively functional in determining CETP and HDL-C levels or a marker for other strongly linked functional variations, we performed extensive genotype and haplotype association analyses of TaqIB and four variants between -3 and +1 kb in the CETP gene that all were in strong linkage disequilibrium. For this analysis, we used the REGRESS cohort consisting of men with angiographically proven CAD, in which we previously observed associations of TaqIB with CETP concentration, HDL-C, progression of atherosclerosis and response to pravastatin treatment (14,27). We hypothesized that in a cluster of tightly linked polymorphisms the gene variation, which is least associated with phenotypic traits, is unlikely to be functional in determining these parameters. Furthermore, haplotype analysis may reveal the effects of single sites within a certain haplotype when the sample size allows comparison of haplotypes that are dissimilar at only one site. Therefore, a novel method that enabled the inclusion of multiple heterozygotes, thereby increasing the sample size and power of the analysis, was used to estimate haplotype effects (28) (M.W.T. Tanck, A.H.E.M. Klerkx, J.W. Jukema, P. De Knijff, J.J.P. Kastelein, A.H. Zwinderman, Ann. Hum. Genet., in press).

Our results show that not the TaqIB polymorphism, but the closely linked -629 promoter polymorphism and a novel promoter polymorphism located at -2708 are independent predictors of the variation in CETP concentration, and involvement of the -971 polymorphism through an interaction effect with -629.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Identification of novel polymorphisms and linkage disequilibrium with TaqIB
The complete DNA sequence between -3060 bp and +1060 bp of the CETP gene was determined in five TaqIB B1B1 and five B2B2 individuals. The consensus of these sequences differed slightly from previously published GenBank sequences (M32992, AF027656, U71187 and U85248), and was therefore submitted to the GenBank under accession number AY172980. Analysis of this DNA sequence revealed the presence 16 single nucleotide polymorphisms (SNPs) including TaqIB, a GAAA repeat of variable length, one variable T-stretch, one deletion mutation, and one combined substitution and deletion (Fig. 1). Of these, eight variations were almost concordant with variation at the TaqIB site. Another three gene variants were less frequent and almost concordant with each other, but not with TaqIB. Finally, five variations were detected only once, and therefore not further studied.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Gene variations in the CETP gene between -3060 and +1060 relative to the transcription start (+1) as detected in five TaqIB B1B1 and five TaqIB B2B2 individuals. Variations are indicated by their position relative to the transcription start with the most frequent nucleotide given left and the least frequent nucleotide given right from the position. The variations that are probably in linkage disequilibrium with the TaqIB polymorphism are given below the schematic representation of the gene and indicated with a double asterisk. Above, the variations that are not linked to TaqIB are given. The variations analyzed in the REGRESS population are in bold. The variations indicated with *, {dagger} and # most likely represent polymorphisms since frequencies of 0.2, 0.2 and 0.33 of the mutant allele were found in 10 individuals. aPositions indicated relative to the transcription start with a GAAA repeat (seeb) of 56 bp between -2020 and -1964. bTetranucleotide GAAA repeat (56). c-971 promoter polymorphism previously investigated by Le Goff et al. (32). d-631 promoter polymorphism (11). e-629 promoter polymorphism previously investigated by Dachet et al. (26).

 
Using the REGRESS population, we further investigated the linkage disequilibrium (LD) between the TaqIB polymorphism and four possibly concordant polymorphisms that may be functional in regulating CETP gene expression. These included the upstream most distant -2708G->A polymorphism and the +784CCC->A polymorphism located 360 bp downstream of TaqIB, of which TRANSFAC analysis indicated that these nucleotide changes might interfere with binding sites of transcription factors involved in regulation of the ApoB and ApoE gene, and therefore may alter CETP gene expression (not shown) (29–31). Furthermore, the previously investigated -629C->A and -971G->A promoter polymorphisms, of which (non)functionality in regulating CETP gene expression in liver cells was already established, were also included (26,32). Allele frequencies of these loci are given in Table 1. All pairwise polymorphism combinations were in significant (P<0.001) LD (Table 1). The strongest pairwise LD was found between the TaqIB polymorphism and the -629 polymorphism (D'=0.98), followed by TaqIB with -2708 and +784. In particular, the least frequent alleles of all polymorphisms were tightly linked to each other. Furthermore, the -971 polymorphism showed the weakest LD with TaqIB and all other variants.


View this table:
[in this window]
[in a new window]
  		Table 1. Allele frequencies and nucleotide changes of five CETP gene variants, and pairwise linkage disequilibrium coefficients between these polymorphisms in the REGRESS population
 
Association of genotypes with CETP concentration and HDL-C
A total of 709 patients were genotyped for all five aforementioned polymorphisms, and in this group baseline CETP concentration and HDL-C were determined in 532 and 702 patients, respectively. Using univariate analysis, these five tightly linked gene variants had similar effects on baseline CETP concentration and HDL-C. Homozygous carriers of the most frequent alleles (denoted as ‘+’) had significantly higher CETP concentration and lower HDL-C levels compared to carriers of the least frequent alleles (denoted as ‘-’), except for -971 and HDL-C (Table 2). These inverse associations with CETP concentration and HDL-C were independent, since baseline CETP concentration and HDL-C were not significantly correlated in the REGRESS population (Pearson r=-0.078, P=0.072). Single polymorphism genotype analysis showed that the -629 promoter polymorphism explained the highest portion of the variance in CETP concentration (7.9%), followed by the distant promoter polymorphism at -2708 (7.0%) and TaqIB (6.0%). The largest portion of variance in HDL-C was explained by -629 (4.6%) and TaqIB (3.9%), whereas -2708 explained only 2.6% of the variance. Multiple polymorphism genotype analysis revealed that all five polymorphisms together in a model with no (= main effects only) and with all possible interactions (= fully saturated model) accounted for 10.3 and 16.2% of the variation in CETP concentration, respectively. For HDL-C, these models explained 5.5 and 9.8% of the total variation, respectively.


View this table:
[in this window]
[in a new window]
  		Table 2. Single polymorphism genotype effects on plasma CETP concentration and HDL-C. Per polymorphism, the mean CETP concentration and HDL-C per genotype, and the percentage of variance explained by the polymorphism (r2) are shown
 
To investigate whether the phenotypic associations of these five polymorphisms, in particular the TaqIB polymorphism, could be attributed to confounding effects of one or more functional polymorphisms, both backward stepwise regression analysis using genotypes and haplotype analyses (see below) were performed. For CETP concentration, the backward regression procedure resulted in a model with independent effects of the -629 (P=0.0025) and -2708 (P=0.025) promoter polymorphisms and baseline triglycerides (P=0.011) as covariate. In the first selection step, the -971 polymorphism was removed, consecutively followed by the TaqIB and the +784 polymorphisms. For HDL-C, this procedure resulted in a model consisting of the -629 genotype (P<0.0001) with baseline triglycerides (P<0.0001) and BMI (P<0.001) as covariates. During the selection steps, the +784 polymorphism was removed first, consecutively followed by TaqIB, -971 and -2708.

Haplotypes and effects on CETP and HDL-C concentration
Haplotypes were assigned to all individuals as described in the Materials and Methods section. The inclusion of both unambiguous and ambiguous individuals greatly increased the number of observations in the data set (397 patients (56%) were heterozygous at two or more loci). Of the 32 possible five polymorphism (5-pol) haplotypes, 20 were estimated to be present. Remarkably, the haplotype consisting of the most frequent alleles of each polymorphism (-2708G/-971G/-629C/TaqIB B1(G)/+784CCC; further referred to as ‘+++++’) and the haplotype consisting of the least frequent alleles (all A; further referred to as ‘-----’) were most prevalent with estimated relative frequencies of 0.4096 and 0.2779, respectively. In accordance with the genotype analyses, the ‘+++++’ haplotype had significantly higher CETP concentration and lower HDL-C compared with the ‘-----’ haplotype. This 5-pol haplotype model explained 16.8% of the variation in CETP concentration and 6.3% of the variation in HDL-C. In Table 3, the effects of the five most frequent 5-pol haplotypes on CETP concentration and HDL-C are shown. Of the 15 remaining haplotypes estimated to be present at a frequency of 0.0258 or below, too few observations were present to observe significant effects. The CETP concentration of the ‘-----’ haplotype was significantly lower than all other compared 5-pol haplotypes. Furthermore, a significant difference in effect on CETP concentration was found for the ‘+-+++’ (-2708G/-971A/-629C/TaqIB B1(G)/+784CCC) and ‘+--+++’ (-2708G/-971A/-629A/TaqIB B1(G)/+784CCC) haplotypes, which only differ at the -629 site. Regarding HDL-C, the ‘-----’ haplotype had significantly higher HDL-C than the ‘+++++’ haplotype and the ‘+-+++’ haplotype. No significant differences in HDL-C were observed between haplotypes differing at a single site. These results did not change when covariates were added to these 5-pol haplotype models, as the estimated effects were essentially the same.


View this table:
[in this window]
[in a new window]
  		Table 3. Effects of the five most common haplotypes on CETP concentration and HDL-C ranked by deviation from the population mean, and P-values for the differences in effects
 
To determine which haplotype model would fit the observed CETP and HDL-C concentrations best, the AIC for all possible ‘one’, two, three, four and five polymorphism haplotype models were calculated. Based on these AIC (Table 4), the best fitting haplotype model for CETP concentrations included the -2708, -971 and -629 polymorphisms, and the next three best scoring models all contained these three polymorphisms. In contrast, the best ‘haplotype’ model for HDL-C only consisted of the -629 polymorphism. Although inclusion of covariates in the haplotype models increased the AIC of the individual models, it did not substantially influence the ranking of the haplotype models.


View this table:
[in this window]
[in a new window]
  		Table 4. The number of haplotypes in the model (#hap), Akaike's information criterion (AIC) values and rank number of the different haplotype models tested on baseline CETP concentration and HDL-C (no covariates). The smallest AIC corresponds to the best fitting model
 
Since the backward stepwise regression analysis indicated independent effects of the -2708 and -629 polymorphisms on CETP concentration, and the haplotype analysis pointed towards an additional contribution of the -971 variation, the effects of these three polymorphisms on CETP concentration were further analyzed. Remarkably, a 2-pol haplotype model that consisted of -971 and -629 best explained the variation in CETP concentration with an r2 of 11.4, whereas the -2708 and -629 model explained 8.2% of the variation in CETP. The effects of the different haplotypes of these models on CETP concentration are depicted in Figure 2. The -2708 and -629 haplotype model confirmed the independent effects on CETP concentration of these polymorphisms observed in the genotype analyses (see Fig. 2A). The most frequent -2708G and -629C alleles acted cumulatively in raising CETP concentration, as was indicated by the intermediate CETP concentration levels of the discordant -2708G/-629A (+-) and -2708A/-629C (-+) haplotypes compared with the -2708G/-629C (++) and -2708A/-629A (--) haplotypes. The difference in CETP concentration was significant for the -2708G/-629A (+-) haplotype compared with ‘++’ and ‘--’ (P=0.001 and P=0.034, respectively). For the -2708A/-629C (-+) the same trend was observed, but this was not significant (versus ‘++’ P=0.072, and versus ‘--’ P=0.052), which could probably be explained by the lower number of observations of this haplotype. Interestingly, in the -971 and -629 model, an interaction effect was observed between the polymorphisms (see Fig. 2C). The least frequent -971A allele in combination with the most frequent -629C allele (-+) resulted in significant raising of CETP concentration compared to the -971C/-629C (++) haplotype (P=0.002), while combination of -971A with the least frequent -629A allele (--) significantly lowered the CETP concentration compared to -971C/-629A (+-) (P=0.002). This interaction effect was confirmed by multiple regression analysis of the -971 and -629 genotypes and their interaction was found to be highly significant (P<0.001) (Fig. 2D). A significant interaction was also observed between the -2708 and -971 polymorphisms (P=0.0468), which for their haplotypes is depicted in Fig. 2B.



View larger version (38K):
[in this window]
[in a new window]
  		Figure 2. 2-Pol haplotype and genotype effects. (A) Effect of haplotype at -2708 and -629 on CETP concentration. (B) Effect of haplotype at -2708 and -971 on CETP concentration. (C) Effect of haplotype at -971 and -629 on CETP concentration. (D) Effect of genotype at -971 and -629 on CETP concentration. (E) Effect of haplotype at -971 and -629 on HDL-C. (F) Effect of haplotype at -629 and TaqIB on HDL-C. Average haplotype effects are depicted with respect to the mean baseline CETP concentration and HDL-C, which were 1.92 µg/ml and 0.92 mmol/l, respectively.

 
Regarding HDL-C, the haplotype analysis confirmed the independent effect of the -629 polymorphism observed in the backward stepwise multiple regression. The second best haplotype model involved the -971 and -629 polymorphisms. However, no significant effect of the -971 polymorphism was found (see Fig. 2E), nor was an interaction effect observed (P=0.191). A possible contribution of the TaqIB polymorphism in determining HDL-C was furthermore indicated by the AIC indices of the TaqIB ‘I’-pol and -629 and TaqIB 2-pol haplotype models. The -629 and TaqIB 2-pol haplotype model revealed intermediate effects of the discordant -629A/TaqIB B1 (-+) and -629C/TaqIB B2 (+-) haplotypes on HDL-C compared with the -629C/TaqIB B1 (++) and -629A/TaqIB B2 (--) haplotypes, but these were not significant (Fig. 2F), nor was a significant interaction effect between these polymorphisms observed (P>0.7).

Furthermore, all other 2-pol and 3-pol haplotype models were analyzed. No significant contributions of either TaqIB or +784 in combination with -629 to CETP concentration or HDL-C were observed. However, TaqIB did show a significant effect on CETP concentration in combination with -2708G (‘++’ versus ‘+-’) and -971A (‘-+’ versus ‘--’). For +784, only the latter was significant.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The CETP TaqIB polymorphism is associated with changes in plasma CETP concentrations, HDL cholesterol, and risk of CAD (11,14,16). Because of its location in intron 1, TaqIB is not expected to directly influence CETP transcriptional regulation or RNA splicing, but more likely serves as a marker for (an)other polymorphism(s) that directly affect CETP gene regulation. We here set out to identify putative functional polymorphisms that are in strong LD with TaqIB, which may account for the observed phenotypic effects. For this, we performed extensive sequence analysis of the CETP gene promoter and intron 1 (-3  and +1 kb relative to the transcriptional start), resulting in the identification of 15 novel polymorphisms. Second, we used extensive genotype analyses and a novel type of haplotype analysis to study the effects of four tightly to TaqIB linked DNA variants on CETP concentration and HDL-C in 709 male patients from the REGRESS cohort.

Polymorphisms and haplotypes
Of the 15 novel CETP gene variations we identified, eight were almost concordant with the TaqIB polymorphism in 10 individuals. Further analysis of the novel -2708 and +784 polymorphisms, and the previously identified TaqIB, -971 and -629 polymorphisms in the REGRESS population, revealed the existence of two completely opposite haplotypes consisting of either the most or least frequent alleles (referred to as ‘+++ ++’ and ‘-----’) at the highest frequencies. Noteworthy, the five most frequent haplotypes accounted for 93% of all existing haplotypes, indicating that TaqIB is part of a limited number of well-conserved haplotypes. The observed concordance of the -971 and -629 polymorphisms and TaqIB is consistent with similar analyses in subsets of the ECTIM study and the OPERA cohort (11,20,32). Others have reported a SNP in intron 7 located more than 8 kb downstream of the currently investigated sequence that was almost concordant with the TaqIB polymorphism in the ECTIM population (10,11). Together, these data suggest that TaqIB may be in tight linkage disequilibrium with an even greater number of nucleotide variations at the CETP gene locus, which, theoretically, could all account for the phenotypic effects observed for the TaqIB polymorphism.

The ‘+++++’ haplotype was associated with significantly lower HDL-C and higher CETP concentration, as was previously observed for both the TaqIB B1 (+) and -629C (+) alleles separately (11,14,25). In the following sections we will discuss the contribution of each of the polymorphisms on CETP concentration and HDL-C starting with the polymorphism that has the largest effect, and we will discuss possible mechanisms that may explain the effects of these gene variations. Finally, we will evaluate the status of the TaqIB polymorphism.

Associations with CETP concentration and mechanism
The proportion of variance explained by the polymorphisms as assessed by univariate analysis (maximally 7.9%) and multivariate analyses (10.3 and 16.2%) indicates that more than one DNA variation and a possible interaction between these variations are responsible for the observed effect on CETP concentration. The -629 variant came forward as the best predictor of variation in CETP concentration in the genotype analysis and was always included in the best fitting haplotype models. Moreover, in every 2-pol haplotype model including -629, significant independent effects of this polymorphism on CETP concentration were observed. Finally, the strongest indication for independent effects of -629 polymorphism was provided by the 5-pol haplotype model, which showed that the mutant -629A (-) allele had a significant CETP concentration lowering effect when the ‘+--++’ and ‘+-+++’ haplotypes were compared.

Surprisingly, the most distant promoter polymorphism located at -2708 was shown to be an independent determinant of CETP concentration in the stepwise regression procedure. From the 5-pol haplotype models, however, no information was obtained on a separate contribution of -2708, which is most likely due to the very strong linkage of the least frequent -2708A allele with all ‘-’ alleles of the other polymorphisms. On the other hand, the independent effect of this site was corroborated by the inclusion of this polymorphism in the best fitting haplotype models, and an independent effect of -2708 in the 2-pol haplotype analysis of -2708 and -629. In this model, the CETP lowering effect of the -2708A allele (-) constituted 72% of the effect observed for -629A (-), suggesting that -2708 and -629 comparably and cumulatively contribute to the variation in CETP concentration.

The -971 polymorphism scored very low in the genotype analyses, which together with its weak linkage disequilibria with all other variants suggested that this gene variation is not important for determining CETP levels. These findings are in agreement with the data of Le Goff and coworkers (32), who observed a significant association of the -971 polymorphism with CETP concentration in the ECTIM population that disappeared upon correction for variation at the -629 and TaqIB sites. They furthermore showed that the -971 site is not functional in regulating CETP gene expression in liver cells (HepG2). Confusingly, -971 was found to be essential in the best scoring haplotype models, and further analysis revealed a strong interaction effect between the -971 and -629 polymorphisms on CETP concentration. A possible explanation for these discrepancies will be discussed below in a mechanistic context.

Our haplotype analyses furthermore indicate that both the TaqIB and +784 polymorphisms do not significantly contribute to the variation in CETP concentration, but that the associations of these polymorphisms observed in the genotype and 2-pol haplotype analyses can be attributed to the linkage disequilibria of these variants with the -2708 and -629 polymorphisms.

All polymorphisms that were found to contribute to variations in CETP concentration are located in the promoter and may therefore be located in cis-acting elements that are functional in regulating the transcriptional activity of the CETP gene. The significant CETP-lowering effect of the -629A allele as observed in our study is in accordance with the in vitro data of Dachet et al. (26). They showed that in HepG2 cells the expression of a reporter gene containing the least frequent -629A allele was 25% lower compared with the most frequent -629C allele. Thus, our finding that the -629 polymorphism accounts for 7.9% of the variation in plasma CETP concentration is likely to be the result of a differential expression of both alleles in the liver, which is the major but not the only site of CETP synthesis (1,33). Since -971 and -629 are in close proximity to each other, one could envision numerous scenarios of interplay between transcription factors binding to these sites that would explain the observed interaction effect. However, Le Goff et al. (32) did not observe specific binding of transcription factors to either variation at -971 in HepG2 cells. The discrepancy between the in vitro data and the association studies might be explained by the fact that the liver is not the only site of CETP synthesis. Adipose tissue was reported to account for a considerable portion of CETP synthesis and in this respect the action of -971 might be adipose tissue-specific, which may result in variations in plasma CETP concentration (33–37). Regarding the -2708 polymorphism, our results indicate that this variation acts independently in determining CETP concentration, which could possibly be explained by the interference of this site with an enhancer or silencer element involved in the transcriptional regulation of the CETP gene. Alternatively, the possibility remains that -2708 and -971 themselves are not functional, but markers for yet other linked functional polymorphism located elsewhere that were not investigated in this study. To determine this, (further) functional analysis of these sites is crucial.

Associations with HDL-C and mechanism
In the univariate analyses, the -629 and TaqIB polymorphisms accounted for 4.6% and 3.9% of the variation in HDL-C, respectively. In the ECTIM study cohort of myocardial infarction (MI) survivors 5.58% of the variation in HDL-C was explained by the -629 polymorphism (11), while the TaqIB polymorphism only accounted for ~1% of the variation in HDL-C in the Framingham Offspring Study (16). This indicates that the contribution of CETP gene variants to HDL-C variation may substantially differ in diseased and healthy populations. In the REGRESS cohort, both stepwise regression genotype analysis and testing of all haplotype models on HDL-C indicated that only the -629 polymorphism was an independent determinant of the variation in HDL-C. This, however, could not be confirmed by comparison of the effects of two haplotypes, which only differed at the -629 site. Although the -2708 polymorphism was shown to play a major role in determining CETP concentration, none of our results pointed towards a contribution of this SNP to HDL-C. With respect to -971, the recent analysis in the ECTIM study showed significant associations of this site with HDL-C, even when corrected for the -629 and TaqIB polymorphisms (32). Our results, in contrast, do not indicate any association of the -971 site with HDL-C. An explanation for this discrepancy will be discussed in a broader context below. Regarding TaqIB, however, we observed a trend towards a possible involvement of this site, as was tentatively inferred by the ranking of the best haplotype models for HDL-C, but this needs further confirmation. Therefore, from these results it can be concluded that of the investigated sites, the -629 polymorphism is the only independent determinant of HDL-C.

The critical role of CETP in HDL metabolism is illustrated by genetic CETP deficiency, in which loss of CETP activity is associated with strongly elevated HDL-C levels (2). The novel CETP inhibitor JTT-705 was furthermore shown to raise HDL-C levels in cholesterol-fed rabbits and humans (38,39). Epidemiological studies, however, consistently indicate that the associations between CETP gene variants and HDL-C are independent of CETP concentration or activity, implying that these associations are not mediated through CETP (10,11,16, this study). This absence of a direct association between CETP concentration and HDL-C levels in epidemiological studies may be related to (genetic) variations in other proteins involved in HDL metabolism (40). In addition, multiple factors are known to have heterogeneous effects on CETP and HDL-C such as age, gender and lifestyle differences in populations of patients and healthy subjects. Together, these factors may be responsible for obscuring the effects of variations in plasma CETP concentration on HDL-C levels. It could, therefore, be argued that finding associations between the -629 polymorphism and both CETP and HDL-C concentration despite these confounding genetic and environmental factors, is indicative of a direct involvement of this site in determining HDL-C levels mediated through variations in CETP concentration. However, the two other polymorphisms that were shown to affect CETP concentration did not influence HDL-C. Moreover, the haplotype analysis suggested a possible role of TaqIB polymorphism for determining HDL-C, which was found to be non-functional in determining CETP concentration. These latter observations rather suggest that the mechanism by which these variations in the CETP gene influence HDL-C does not involve CETP. Le Goff et al. (32) propose the existence of another as yet unidentified polymorphism linked to -971, -629 and TaqIB, which simultaneously modulates HDL-C and CETP concentration. We here show that, of the five investigated polymorphisms, the -629 site is the only polymorphism that is independently associated with HDL-C, but the mechanism that can explain these associations remains unclear.

Concluding remarks
As mentioned above, in contrast to the data published by Le Goff and co-workers (32), we did not find an association between -971 and HDL-C. This inconsistency may be explained by the use of a mixed population of women, men, cases and controls selected from the ECTIM cohort, whereas the REGRESS population consisted of only male CAD cases. Accordingly, HDL-C levels differed largely between these two populations. However, this discrepancy and other inconsistent observations in our and other cohorts with respect to associations of TaqIB and -629 with CETP concentration or HDL-C (17,19,20,32), teach us that it may prove very difficult to draw general conclusions regarding the value of CETP gene variations while using epidemiological data. We are convinced, however, that the use of well-characterized homogenous populations is imperative for identifying putative functional polymorphisms, as the influence of (unknown) confounding factors is in all probability reduced.

To the best of our knowledge, this is one of the first studies investigating phenotypic associations of a cluster of tightly linked polymorphisms using haplotype analyses in unrelated subjects. Our data clearly subscribe that a detailed analysis of a larger number of polymorphisms is necessary to appoint functional variants (41,42). More specifically, our analyses show that 2-pol haplotype models could only be used in combination with larger haplotype models to reveal the complex associations between CETP gene polymorphisms and CETP concentration. This is for instance illustrated by the significant effect of TaqIB on CETP concentration observed in the 2-pol haplotype models consisting of TaqIB and either -2708 or -971. Neither in the stepwise multiple regression procedure of five polymorphisms nor in the comparison of the haplotype models was TaqIB indicated to have an independent effect on CETP concentration. This allowed us to conclude that in the 2-pol haplotype models TaqIB probably served as a substitute for -629. We did not include all polymorphisms of the -629/TaqIB cluster in our analysis for practical reasons. If reporter gene analyses show that -2708 and -971 do not directly influence CETP gene expression, our haplotype analysis needs to be extended to additional polymorphisms that are in LD. Nevertheless, this approach handed us new tools to unravel the complexity of the relationship between CETP gene polymorphisms and CETP concentration and HDL-C.

It is not clear whether the effect of these polymorphisms on CETP concentration and HDL-C can be extrapolated to the previously observed associations of TaqIB with angiographically assessed progression of CAD and response to pravastatin (14). We performed similar analysis of genotype effects of these five polymorphisms on mean segment diameter (MSD) and minimum obstruction diameter (MOD) as previously done for TaqIB. These analyses indicated that both -629 and TaqIB had equal effects on progression, and that there was an interaction effect between treatment and genotype at these sites (not shown). The power of subsequent 5-pol stepwise regression and haplotype analyses, however, was not sufficient to resolve whether these associations can be attributed to either -629 or TaqIB (not shown). In the Reykjavik Study, TaqIB explained the risk of MI better than -629 genotype (43), and it was suggested that the associations of TaqIB with the development of MI are due to linkage disequilibrium of TaqIB to yet another unknown polymorphism. We recently established, however, that baseline CETP concentration was more indicative for the clinical outcome of the REGRESS trial than genetic variation at the TaqIB site (A.H.E.M. Klerkx, G.J. de Grooth, A.H. Zwinderman, J.W. Jukema, J.A. Kuivenhoven, J.J.P. Kastelein, manuscript in preparation). From this, it can be hypothesized that, with regard to clinical outcome, variations in the CETP gene may merely be markers for plasma CETP concentration levels.

In conclusion, we here show that variation at the TaqIB site does not play a role in determining either CETP concentration or HDL-C in the REGRESS population, but instead has served as a useful marker for the effects of tightly linked polymorphisms. Using extensive genotype and haplotype analysis, we have demonstrated that in a cluster of five tightly linked polymorphisms, the -629 promoter polymorphism has independent effects on CETP concentration and HDL-C. Secondly, variations located at -2708 and at -971 in the CETP promoter were identified as putatively functional sites that may act together with -629 in determining CETP concentration in an independent cumulative and interactive manner, respectively. Further in vitro functional analysis of these polymorphisms will be critical to provide better insight into the complex regulation of CETP gene expression, and will help to understand the mechanisms that underlie the observed phenotype–genotype associations.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Study subjects
Subjects were participants in the Regress Growth and Evaluation Statin Study (REGRESS), which was designed to assess the effect of pravastatin on progression of coronary atherosclerosis (27). In short, all patients were Dutch Caucasians males less than 70 years of age who presented an obstruction of at least 50% of a major coronary artery as assessed by quantitative coronary angiography, and had total cholesterol levels between 4 and 8 mmol/l and plasma triglyceride levels under 4 mmol/l. Patients were treated with either pravastatin (40 mg/day) or placebo for 2 years. All participants gave informed consent.

DNA sequence analysis
Four subjects with the TaqIB B1B1 genotype (presence of the TaqIB restriction site) with relatively high CETP and low HDL-C, and four TaqIB B2B2 subjects with relatively low CETP and high HDL-C levels were selected from the REGRESS study for DNA sequence analysis. In addition, the corresponding DNA sequence of three human genomic library clones (Max–Planck Institute for molecular genetics Berlin, Germany) was determined, namely a YAC clone (ICRFy900G05108D1) and two PAC clones (RPCIP704H20548Q2 and RPCIP704N05733Q2). Standard methods were used for the extraction of genomic DNA from whole blood and plasmid DNA isolation from the YAC and PAC cultures.

To determine the DNA sequence of the region between -3028 and +1060 of the CETP gene, primers were based on GenBank sequences M32992, AF027656, U71187 and U85248, generating overlapping PCR fragments of approximately 1000 bp. Both strands of the generated PCR products were sequenced with an end primer and an additional internal primer on an automated ABI system using Big Dye Terminator (Applied Biosystems). DNA sequence data were analyzed using GCG (Genetics Computer Group Inc., Madison, WI, USA) and AutoAssembler (Applied Biosystems) software.

Genotype analysis
Genotyping was performed using PCR based restriction fragment length polymorphism analysis. When no disappearance or creation of a restriction site occurred as a result of a newly identified sequence variation, we designed mismatch primers creating forced restriction sites. Table 5 gives the PCR primers, the restriction enzymes used to detect the investigated polymorphisms, and the length of the resulting fragments in the homozygote most and least frequent genotypes. A typical 25 µl PCR reaction mix consisted of 0.1–0.5 µg genomic DNA, 100 nmol/l of each primer, 100 µmol/l of dNTPs and 0.5 units SuperTaq polymerase (HT Biotechnology Ltd, Cambridge, UK) in 1x SuperTaq buffer (10 mmol/l Tris–HCl, pH 9.0, 1.5 mmol/l MgCl2, 50 mmol/l KCl, 0.1% Triton X-100 and 0.01% (w/v) stabilizer). After initial denaturation of the DNA at 94°C for 5 min, the PCR conditions were 94°C for 1 min, 62°C for 1 min and 72°C for 30 s for 30 cycles, followed by 10 min at 72°C. Ten microliters of the PCR sample were digested with two units of the appropriate restriction enzyme following the manufacturer's instructions in a final reaction volume of 20 µl. The restriction fragments were separated by gel electrophoresis on a 2 or 3% agarose gel containing ethidiumbromide and visualized on a UV transilluminator.


View this table:
[in this window]
[in a new window]
  		Table 5. Tools for genotype analysis of five polymorphisms in the CETP gene; primers and restriction enzymes used for detection of the sequence variations
 
Biochemical analysis
The assays for measuring lipid and lipoprotein parameters and a detailed analysis of the lipid variables in the REGRESS study have already been reported elsewhere (14). Baseline CETP concentration was determined using a two-antibody sandwich immunoassay, which was developed by Niemeijer-Kanters et al. (44). As a standard, pool plasma containing 2 µg CETP/ml was taken.

Statistical analysis
Genotype associations-Single CETP polymorphism genotype effects on baseline plasma CETP and HDL-C concentration were tested by analysis of variance adjusted for covariates. Multiple CETP polymorphism genotype effects were investigated by a backward regression analysis (P-stay: 0.05) with all five CETP polymorphisms and the covariates in the initial model. Baseline plasma triglyceride concentration was included in the models for the CETP concentration analyses. For the HDL-C concentration analyses, smoking, body mass index and plasma triglyceride concentration were included as covariates. In all analyses, the proportion of variance attributable to the polymorphisms (r2) was calculated as the ratio of the sum of squares due to genotype effects to the total sum of squares. Throughout, P-values<0.05 were interpreted as significant. All statistical analyses were done with SAS software (version 8, SAS Institute, Cary, NC, USA).

Haplotype associations.
Analysis of Hardy–Weinberg equilibrium was performed according to the procedure described by Guo and Thompson (45) and linkage disequilibria between pairs of polymorphic positions were calculated according to the likelihood-ratio test (46). Both procedures are incorporated in the Arlequin 2.000 software package (47). The extent of disequilibrium was expressed in terms of D'=D/Dmax or D/Dmin (48).

The haplotype effects were estimated using a method described by Tanck et al. (M.W.T. Tanck, A.H.E.M. Klerkx, J.W. Jukema, P. De Knijff, J.J.P. Kastelein, A.H. Zwinderman, Ann. Hum. Genet., in press). In short, haplotypes were assigned to all unambiguous individuals, i.e. those who were homozygous at all polymorphisms or heterozygous at a single polymorphism. For the remaining ambiguous individuals, the haplotype pairs compatible with their genotype were determined and the posterior probabilities were calculated using Bayes' theorem and the maximum likelihood relative haplotype frequencies, which were estimated using the genotype frequencies and an expectation-maximization (E-M) algorithm (49). These posterior probabilities were used as weights in the statistical model.

The statistical model was basically a linear regression model in which the expected CETP or HDL-C concentration in a subject, y, was assumed to be a linear function of haplotype effects (and effects of confounders): y=ß1X1+ß2X2 +···+ ßmXm +···, where Xl (l=1,... , m) attained values 0, 1 or 2 denoting presence of 0, 1 or 2 copies of haplotype l in this subject (m=number of haplotypes estimated to be present in the population). The regression parameter ßl represents the average effect of haplotype l on CETP or HDL-C concentration, which under Hardy–Weinberg genotype frequencies is identical to the average excess (50). Since m was possibly very large, and some haplotypes occurred with very low frequency, the estimated effects of less frequent haplotypes were expected to show a huge variation. To circumvent this problem, we used a penalized log-likelihood approach. Comparable to the ‘molecular similarity’ approach described by Thomas et al. (51), we assumed that haplotypes sharing the same alleles would show a similar effect and, secondly, that the extent of this similarity would be related to the number of alleles shared. This way, information available for similar but more frequent haplotypes could be included in the estimation of the effect of the less frequent haplotype, leading to estimates that are more reliable. These assumptions were incorporated in a weighted log-likelihood model by introducing a penalty term (52), where differences in effects of similar haplotypes were penalized. The optimal magnitude of the penalty term was determined through cross-validation using the generalized cross-validation criterion (53).

The parameters were estimated using an E-M algorithm. In the E-step, the posterior probabilities of the haplotype pairs were calculated based on the phenotype of the individual subject (in the first E-step, the initial probabilities calculated using the estimated haplotype frequencies were used). In the M-step, the haplotype effects were estimated by solving the first derivative of the penalized log-likelihood function. These two steps were alternated until convergence. The variance of the haplotype effects was estimated using Louis' method (54).

Next to the five-polymorphism haplotype effects, effects of all possible ‘one’, two, three and four polymorphism haplotypes were also estimated using the procedure described above. For each haplotype model, the Akaike Information Criterion (AIC) (55) was calculated. Although the penalty had little influence on the ranking of the haplotype models based on the AIC, the reported AICs were calculated in models in which the penalty term was set to zero.


   ACKNOWLEDGEMENTS
 
The authors are grateful to M. Eikenboom, L. Klaaijsen and J. Stallen for technical assistance. This work supported by Seed Capital Investments (SCI-II, The Netherlands) and the Netherlands Heart Foundation grant numbers 2000.073 and 2000.125.


   FOOTNOTES
 
* To whom correspondence should be addressed at: Department of Vascular Medicine, G1-114, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. Tel: +31 205666118; Fax: +31 205669232; Email: a.klerkx{at}amc.uva.nl Back

The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Tall, A.R. (1993) Plasma cholesteryl ester transfer protein. J. Lipid Res., 34, 1255–1274.[Web of Science][Medline]

  2. Inazu, A., Jiang, X.C., Haraki, T., Yagi, K., Kamon, N., Koizumi, J., Mabuchi, H., Takeda, R., Takata, K. and Moriyama, Y. (1994) Genetic cholesteryl ester transfer protein deficiency caused by two prevalent mutations as a major determinant of increased levels of high density lipoprotein cholesterol. J. Clin. Invest., 94, 1872–1882.[Web of Science][Medline]

  3. Zhong, S., Sharp, D.S., Grove, J.S., Bruce, C., Yano, K., Curb, J.D. and Tall, A.R. (1996) Increased coronary heart disease in Japanese–American men with mutation in the cholesteryl ester transfer protein gene despite increased HDL levels. J. Clin. Invest., 97, 2917–2923.[Web of Science][Medline]

  4. Hirano, K., Yamashita, S. and Matsuzawa, Y. (2000) Pros and cons of inhibiting cholesteryl ester transfer protein. Curr. Opin. Lipidol., 11, 589–596.[CrossRef][Web of Science][Medline]

  5. Manninen, V., Elo, M.O., Frick, M.H., Haapa, K., Heinonen, O.P., Heinsalmi, P., Helo, P., Huttunen, J.K., Kaitaniemi, P., Koskinen, P. et al. (1988) Lipid alterations and decline in the incidence of coronary heart disease in the Helsinki Heart Study. JAMA, 260, 641–651.[Abstract/Free Full Text]

  6. Bruce, C., Chouinard, R.A., Jr and Tall, A.R. (1998) Plasma lipid transfer proteins, high-density lipoproteins, and reverse cholesterol transport. A. Rev. Nutr., 18, 297–330.

  7. Yamashita, S., Hirano, K., Sakai, N. and Matsuzawa, Y. (2000) Molecular biology and pathophysiological aspects of plasma cholesteryl ester transfer protein. Biochim. Biophys. Acta, 1529, 257–275.[Medline]

  8. Drayna, D. and Lawn, R. (1987) Multiple RFLPs at the human cholesteryl ester transfer protein (CETP) locus. Nucl. Acids Res., 15, 4698.[Free Full Text]

  9. Funke, H., Wiebusch, H., Fuer, L., Muntoni, S., Schulte, H. and Assmann, G. (1994) Identification of mutations in the cholesteryl ester transfer protein in Europeans with elevated high density lipoprotein cholesterol. Circulation, 90, I-241.[Abstract/Free Full Text]

  10. Fumeron, F., Betoulle, D., Luc, G., Behague, I., Ricard, S., Poirier, O., Jemaa, R., Evans, A., Arveiler, D. and Marques-Vidal, P. (1995) Alcohol intake modulates the effect of a polymorphism of the cholesteryl ester transfer protein gene on plasma high density lipoprotein and the risk of myocardial infarction. J. Clin. Invest., 96, 1664–1671.[Web of Science][Medline]

  11. Corbex, M., Poirier, O., Fumeron, F., Betoulle, D., Evans, A., Ruidavets, J.B., Arveiler, D., Luc, G., Tiret, L. and Cambien, F. (2000) Extensive association analysis between the CETP gene and coronary heart disease phenotypes reveals several putative functional polymorphisms and gene–environment interaction. Genet. Epidemiol., 19, 64–80.[CrossRef][Web of Science][Medline]

  12. Freeman, D.J., Griffin, B.A., Holmes, A.P., Lindsay, G.M., Gaffney, D., Packard, C.J. and Shepherd, J. (1994) Regulation of plasma HDL cholesterol and subfraction distribution by genetic and environmental factors. Associations between the TaqI B RFLP in the CETP gene and smoking and obesity. Arterioscler. Thromb., 14, 336–344.[Abstract/Free Full Text]

  13. Bernard, S., Moulin, P., Lagrost, L., Picard, S., Elchebly, M., Ponsin, G., Chapuis, F. and Berthezene, F. (1998) Association between plasma HDL-cholesterol concentration and Taq1B CETP gene polymorphism in non-insulin-dependent diabetes mellitus. J. Lipid Res., 39, 59–65.[Abstract/Free Full Text]

  14. Kuivenhoven, J.A., Jukema, J.W., Zwinderman, A.H., de Knijff, P., McPherson, R., Bruschke, A.V., Lie, K.I. and Kastelein, J.J.P. (1998) The role of a common variant of the cholesteryl ester transfer protein gene in the progression of coronary atherosclerosis. The Regression Growth Evaluation Statin Study Group. New Engl. J. Med., 338, 86–93.[Abstract/Free Full Text]

  15. Gudnason, V., Kakko, S., Nicaud, V., Savolainen, M.J., Kesaniemi, Y.A., Tahvanainen, E. and Humphries, S. (1999) Cholesteryl ester transfer protein gene effect on CETP activity and plasma high-density lipoprotein in European populations. The EARS Group. Eur. J. Clin. Invest., 29, 116–128.[CrossRef][Web of Science][Medline]

  16. Ordovas, J.M., Cupples, L.A., Corella, D., Otvos, J.D., Osgood, D., Martinez, A., Lahoz, C., Coltell, O., Wilson, P.W. and Schaefer, E.J. (2000) Association of cholesteryl ester transfer protein-TaqIB polymorphism with variations in lipoprotein subclasses and coronary heart disease risk: the Framingham study. Arterioscler. Thromb. Vasc. Biol., 20, 1323–1329.[Abstract/Free Full Text]

  17. Inazu, A. and Mabuchi, H. (2001) An unsolved mystery of promoter variation in CETP gene and atherosclerosis. Eur. J. Clin. Invest., 31, 558–559.[CrossRef][Web of Science][Medline]

  18. Corella, D., Saiz, C., Guillen, M., Portoles, O., Mulet, F., Gonzalez, J.I. and Ordovas, J.M. (2000) Association of TaqIB polymorphism in the cholesteryl ester transfer protein gene with plasma lipid levels in a healthy spanish population. Atherosclerosis, 152, 367–376.[CrossRef][Web of Science][Medline]

  19. Meguro, S., Takei, I., Murata, M., Hirose, H., Takei, N., Mitsuyoshi, Y., Ishii, K., Oguchi, S., Shinohara, J., Takeshita, E. et al. (2001) Cholesteryl ester transfer protein polymorphism associated with macroangiopathy in Japanese patients with type 2 diabetes. Atherosclerosis, 156, 151–156.[CrossRef][Web of Science][Medline]

  20. Kakko, S., Tamminen, M., Paivansalo, M., Kauma, H., Rantala, A.O., Lilja, M., Reunanen, A., Kesaniemi, Y.A. and Savolainen, M.J. (2001) Variation at the cholesteryl ester transfer protein gene in relation to plasma high density lipoproteins cholesterol levels and carotid intima-media thickness. Eur. J. Clin. Invest., 31, 593–602.[CrossRef][Web of Science][Medline]

  21. Goto, A., Sasai, K., Suzuki, S., Fukutomi, T., Ito, S., Matsushita, T., Okamoto, M., Suzuki, T., Itoh, M., Okumura-Noji, K. et al. (2001) Cholesteryl ester transfer protein and atherosclerosis in Japanese subjects: a study based on coronary angiography. Atherosclerosis, 159, 153–163.[CrossRef][Web of Science][Medline]

  22. Hannuksela, M.L., Liinamaa, M.J., Kesaniemi, Y.A. and Savolainen, M.J. (1994) Relation of polymorphisms in the cholesteryl ester transfer protein gene to transfer protein activity and plasma lipoprotein levels in alcohol drinkers. Atherosclerosis, 110, 35–44.[CrossRef][Web of Science][Medline]

  23. Vohl, M.C., Lamarche, B., Pascot, A., Leroux, G., Prud'homme, D., Bouchard, C., Nadeau, A. and Despres, J.P. (1999) Contribution of the cholesteryl ester transfer protein gene TaqIB polymorphism to the reduced plasma HDL-cholesterol levels found in abdominal obese men with the features of the insulin resistance syndrome. Int. J. Obes. Relat Metab Disord., 23, 918–925.[CrossRef][Web of Science][Medline]

  24. Brousseau, M.E., O'Connor, J.J., Jr, Ordovas, J.M., Collins, D., Otvos, J.D., Massov, T., McNamara, J.R., Rubins, H.B., Robins, S.J. and Schaefer, E.J. (2002) Cholesteryl ester transfer protein TaqI B2B2 genotype is associated with higher HDL cholesterol levels and lower risk of coronary heart disease end points in men with HDL deficiency: Veterans Affairs HDL Cholesterol Intervention Trial. Arterioscler. Thromb. Vasc. Biol., 22, 1148–1154.[Abstract/Free Full Text]

  25. Kuivenhoven, J.A., de Knijff, P., Boer, J.M., Smalheer, H.A., Botma, G.J., Seidell, J.C., Kastelein, J.J. and Pritchard, P.H. (1997) Heterogeneity at the CETP gene locus. Influence on plasma CETP concentrations and HDL cholesterol levels. Arterioscler. Thromb. Vasc. Biol., 17, 560–568.[Abstract/Free Full Text]

  26. Dachet, C., Poirier, O., Cambien, F., Chapman, J. and Rouis, M. (2000) New functional promoter polymorphism, CETP/-629, in cholesteryl ester transfer protein (CETP) gene related to CETP mass and high density lipoprotein cholesterol levels: role of Sp1/Sp3 in transcriptional regulation. Arterioscler. Thromb. Vasc. Biol., 20, 507–515.[Abstract/Free Full Text]

  27. Jukema, J.W., Bruschke, A.V., van Boven, A.J., Reiber, J.H., Bal, E.T., Zwinderman, A.H., Jansen, H., Boerma, G.J., van Rappard, F.M. and Lie, K.I. (1995) Effects of lipid lowering by pravastatin on progression and regression of coronary artery disease in symptomatic men with normal to moderately elevated serum cholesterol levels. The Regression Growth Evaluation Statin Study (REGRESS). Circulation, 91, 2528–2540.[Abstract/Free Full Text]

  28. Tanck, M.W.T., Jukema, J.W., Klerkx, A.H.E.M., Kuivenhoven, J.A., de Knijff, P. and Zwinderman, A.H. (2001) A novel method to estimate haplotype effects in patient populations. Circulation, 104 (Suppl. II), 828.

  29. Chang, D.J., Paik, Y.K., Leren, T.P., Walker, D.W., Howlett, G.J. and Taylor, J.M. (1990) Characterization of a human apolipoprotein E gene enhancer element and its associated protein factors. J. Biol. Chem., 265, 9496–9504.[Abstract/Free Full Text]

  30. Paulweber, B., Onasch, M.A., Nagy, B.P. and Levy-Wilson, B. (1991) Similarities and differences in the function of regulatory elements at the 5' end of the human apolipoprotein B gene in cultured hepatoma (HepG2) and colon carcinoma (CaCo-2) cells. J. Biol. Chem., 266, 24149–24160.[Abstract/Free Full Text]

  31. Heinemeyer, T., Wingender, E., Reuter, I., Hermjakob, H., Kel, A.E., Kel, O.V., Ignatieva, E.V., Ananko, E.A., Podkolodnaya, O.A., Kolpakov, F.A. et al. (1998) Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucl. Acids Res., 26, 362–367.[Abstract/Free Full Text]

  32. Le Goff, W., Guerin, M., Nicaud, V., Dachet, C., Luc, G., Arveiler, D., Ruidavets, J.B., Evans, A., Kee, F., Morrison, C. et al. (2002) A novel cholesteryl ester transfer protein promoter polymorphism (-971G/A) associated with plasma high-density lipoprotein cholesterol levels. Interaction with the TaqIB and -629C/A polymorphisms. Atherosclerosis, 161, 269–279.[CrossRef][Web of Science][Medline]

  33. Jiang, X.C., Moulin, P., Quinet, E., Goldberg, I.J., Yacoub, L.K., Agellon, L.B., Compton, D., Schnitzer-Polokoff, R. and Tall, A.R. (1991) Mammalian adipose tissue and muscle are major sources of lipid transfer protein mRNA. J. Biol. Chem., 266, 4631–4639.[Abstract/Free Full Text]

  34. Richardson, M.A., Berg, D.T., Johnston, P.A., McClure, D. and Grinnell, B.W. (1996) Human liposarcoma cell line, SW872, secretes cholesteryl ester transfer protein in response to cholesterol. J. Lipid Res., 37, 1162–1166.[Abstract]

  35. Gauthier, B., Robb, M., Gaudet, F., Ginsburg, G.S. and McPherson, R. (1999) Characterization of a cholesterol response element (CRE) in the promoter of the cholesteryl ester transfer protein gene: functional role of the transcription factors SREBP-1a, -2, and YY1. J. Lipid Res., 40, 1284–1293.[Abstract/Free Full Text]

  36. Luo, Y. and Tall, A.R. (2000) Sterol upregulation of human CETP expression in vitro and in transgenic mice by an LXR element. J. Clin. Invest., 105, 513–520.[Web of Science][Medline]

  37. Izem, L. and Morton, R.E. (2001) Cholesteryl ester transfer protein biosynthesis and cellular cholesterol homeostasis are tightly interconnected. J. Biol. Chem., 276, 26534–26541.[Abstract/Free Full Text]

  38. Okamoto, H., Yonemori, F., Wakitani, K., Minowa, T., Maeda, K. and Shinkai, H. (2000) A cholesteryl ester transfer protein inhibitor attenuates atherosclerosis in rabbits. Nature, 406, 203–207.[CrossRef][Medline]

  39. de Grooth, G.J., Kuivenhoven, J.A., Stalenhoef, A.F., de Graaf, J., Zwinderman, A.H., Posma, J.L., van Tol, A. and Kastelein, J.J. (2002) Efficacy and safety of a novel cholesteryl ester transfer protein inhibitor, JTT-705, in humans: a randomized phase II dose-response study. Circulation, 105, 2159–2165.[Abstract/Free Full Text]

  40. Von Eckardstein, A., Nofer, J.R. and Assmann, G. (2001) High density lipoproteins and arteriosclerosis. Role of cholesterol efflux and reverse cholesterol transport. Arterioscler. Thromb. Vasc. Biol., 21, 13–27.[Abstract/Free Full Text]

  41. Villard, E., Tiret, L., Visvikis, S., Rakotovao, R., Cambien, F. and Soubrier, F. (1996) Identification of new polymorphisms of the angiotensin I-converting enzyme (ACE) gene, and study of their relationship to plasma ACE levels by two-QTL segregation-linkage analysis. Am. J. Hum. Genet., 58, 1268–1278.[Web of Science][Medline]

  42. Tregouet, D.A., Barbaux, S., Escolano, S., Tahri, N., Golmard, J.L., Tiret, L. and Cambien, F. (2002) Specific haplotypes of the P-selectin gene are associated with myocardial infarction. Hum. Mol. Genet., 11, 2015–2023.[Abstract/Free Full Text]

  43. Eiriksdottir, G., Bolla, M.K., Thorsson, B., Sigurdsson, G., Humphries, S.E. and Gudnason, V. (2001) The -629C>A polymorphism in the CETP gene does not explain the association of TaqIB polymorphism with risk and age of myocardial infarction in Icelandic men. Atherosclerosis, 159, 187–192.[CrossRef][Web of Science][Medline]

  44. Niemeijer-Kanters, S.D., Dallinga-Thie, G.M., Ruijter-Heijstek, F.C., Algra, A., Erkelens, D.W., Banga, J.D. and Jansen, H. (2001) Effect of intensive lipid-lowering strategy on low-density lipoprotein particle size in patients with type 2 diabetes mellitus. Atherosclerosis, 156, 209–216.[CrossRef][Web of Science][Medline]

  45. Guo, S.W. and Thompson, E.A. (1992) Performing the exact test of Hardy–Weinberg proportion for multiple alleles. Biometrics, 48, 361–372.[CrossRef][Web of Science][Medline]

  46. Slatkin, M. and Excoffier, L. (1996) Testing for linkage disequilibrium in genotypic data using the Expectation-Maximization algorithm. Heredity, 76 (Pt 4), 377–383.

  47. Schneider, S., Roessli, D., and Excoffier, L. (2000) Arlequin ver. 2.000: a Software for Population Genetics Data Analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland.

  48. Thompson, E.A., Deeb, S., Walker, D. and Motulsky, A.G. (1988) The detection of linkage disequilibrium between closely linked markers: RFLPs at the AI-CIII apolipoprotein genes. Am. J. Hum. Genet., 42, 113–124.[Web of Science][Medline]

  49. Excoffier, L. and Slatkin, M. (1995) Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol. Biol. Evol., 12, 921–927.[Abstract]

  50. Templeton, A.R. (1987) The general relationship between average effect and average excess. Genet. Res., 49, 69–70.[Web of Science][Medline]

  51. Thomas, D.C., Morrison, J.L. and Clayton, D.G. (2001) Bayes estimates of haplotype effects. Genet. Epidemiol., 21 (Suppl. 1), S712–S717.

  52. Eilers, P.H.C. and Marx, B.D. (1996) Flexible smoothing with B-splines and penalties. Stat. Sci., 11, 89–102.[CrossRef][Web of Science]

  53. Golub, G.H., Heath, M. and Wahba, G. (1979) Generalized cross-validation as a method for choosing a good ridge parameter. Technometrics, 21, 215–223.[CrossRef][Web of Science]

  54. Louis, T.A. (1982) Finding the observed information matrix when using the EM algorithm. J. Stat. Soc. Ser. B, 44, 226–233.

  55. Akaike, H. (1973) Information theory and an extension of the maximum likelihood principle. Proceeding of 2nd International Symposium on Information Theory, Suppl. Problems of Control and Information Theory, 267–281.

  56. Talmud, P.J., Edwards, K.L., Turner, C.M., Newman, B., Palmen, J.M., Humphries, S.E. and Austin, M.A. (2000) Linkage of the cholesteryl ester transfer protein (CETP) gene to LDL particle size: use of a novel tetranucleotide repeat within the CETP promoter. Circulation, 101, 2461–2466.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Circ Cardiovasc GenetHome page
P. M. Ridker, G. Pare, A. N. Parker, R. Y.L. Zee, J. P. Miletich, and D. I. Chasman
Polymorphism in the CETP Gene Region, HDL Cholesterol, and Risk of Future Myocardial Infarction: Genomewide Analysis Among 18 245 Initially Healthy Women From the Women's Genome Health Study
Circ Cardiovasc Genet, February 1, 2009; 2(1): 26 - 33.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
J. J. Regieli, J. W. Jukema, D. E. Grobbee, J. J.P. Kastelein, J. A. Kuivenhoven, A. H. Zwinderman, Y. van der Graaf, M. L. Bots, and P. A. Doevendans
CETP genotype predicts increased mortality in statin-treated men with proven cardiovascular disease: an adverse pharmacogenetic interaction
Eur. Heart J., November 2, 2008; 29(22): 2792 - 2799.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
T. Y Li, C. Zhang, F. W Asselbergs, L. Qi, E. Rimm, D. J Hunter, and F. B Hu
Interaction between dietary fat intake and the cholesterol ester transfer protein TaqIB polymorphism in relation to HDL-cholesterol concentrations among US diabetic men
Am. J. Clinical Nutrition, November 1, 2007; 86(5): 1524 - 1529.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
N. Spielmann, A. S. Leon, D. C. Rao, T. Rice, J. S. Skinner, C. Bouchard, and T. Rankinen
CETP genotypes and HDL-cholesterol phenotypes in the HERITAGE Family Study
Physiol Genomics, September 11, 2007; 31(1): 25 - 31.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
B. D. Horne, N. J. Camp, J. L. Anderson, C. P. Mower, J. L. Clarke, M. J. Kolek, J. F. Carlquist, and for the Intermountain Heart Collaborative Study Gr
Multiple Less Common Genetic Variants Explain the Association of the Cholesteryl Ester Transfer Protein Gene With Coronary Artery Disease
J. Am. Coll. Cardiol., May 22, 2007; 49(20): 2053 - 2060.
[Abstract] [Full Text] [PDF]

Home page
 JAMAHome page
T. M. Morgan, H. M. Krumholz, R. P. Lifton, and J. A. Spertus
Nonvalidation of Reported Genetic Risk Factors for Acute Coronary Syndrome in a Large-Scale Replication Study
JAMA, April 11, 2007; 297(14): 1551 - 1561.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
W. A. van der Steeg, G. K. Hovingh, A. H. E. M. Klerkx, B. A. Hutten, I. C. Nootenboom, J. H. M. Levels, A. van Tol, G. M. Dallinga-Thie, A. H. Zwinderman, J. J. P. Kastelein, et al.
Cholesteryl ester transfer protein and hyperalphalipoproteinemia in Caucasians
J. Lipid Res., March 1, 2007; 48(3): 674 - 682.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
J. F. Thompson, L. S. Wood, E. H. Pickering, B. DeChairo, and C. L. Hyde
High-density genotyping and functional SNP localization in the CETP gene
J. Lipid Res., February 1, 2007; 48(2): 434 - 443.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
K. L.E. Klos, C. F. Sing, E. Boerwinkle, S. C. Hamon, T. J. Rea, A. Clark, M. Fornage, and J. E. Hixson
Consistent Effects of Genes Involved in Reverse Cholesterol Transport on Plasma Lipid and Apolipoprotein Levels in CARDIA Participants
Arterioscler Thromb Vasc Biol, August 1, 2006; 26(8): 1828 - 1836.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
E. Frisdal, A. H.E.M. Klerkx, W. L. Goff, M. W.T. Tanck, J.-P. Lagarde, J. W. Jukema, J. J.P. Kastelein, M. J. Chapman, and M. Guerin
Functional interaction between -629C/A, -971G/A and -1337C/T polymorphisms in the CETP gene is a major determinant of promoter activity and plasma CETP concentration in the REGRESS Study
Hum. Mol. Genet., September 15, 2005; 14(18): 2607 - 2618.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
H. Deguchi, N. M. Pecheniuk, D. J. Elias, P. M. Averell, and J. H. Griffin
High-Density Lipoprotein Deficiency and Dyslipoproteinemia Associated With Venous Thrombosis in Men
Circulation, August 9, 2005; 112(6): 893 - 899.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
S. E. Borggreve, H. L. Hillege, B. H. R. Wolffenbuttel, P. E. de Jong, S. J. L. Bakker, G. van der Steege, A. van Tol, R. P. F. Dullaart, and on behalf of the PREVEND Study Group
The Effect of Cholesteryl Ester Transfer Protein -629C->A Promoter Polymorphism on High-Density Lipoprotein Cholesterol Is Dependent on Serum Triglycerides
J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4198 - 4204.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
S.M. Boekholdt, F.M. Sacks, J.W. Jukema, J. Shepherd, D.J. Freeman, A.D. McMahon, F. Cambien, V. Nicaud, G.J. de Grooth, P.J. Talmud, et al.
Cholesteryl Ester Transfer Protein TaqIB Variant, High-Density Lipoprotein Cholesterol Levels, Cardiovascular Risk, and Efficacy of Pravastatin Treatment: Individual Patient Meta-Analysis of 13 677 Subjects
Circulation, January 25, 2005; 111(3): 278 - 287.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
G. J. de Grooth, A. H. E. M. Klerkx, E. S. G. Stroes, A. F. H. Stalenhoef, J. J. P. Kastelein, and J. A. Kuivenhoven
A review of CETP and its relation to atherosclerosis
J. Lipid Res., November 1, 2004; 45(11): 1967 - 1974.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
D.-A. Tregouet, S. Ricard, V. Nicaud, I. Arnould, S. Soubigou, M. Rosier, N. Duverger, O. Poirier, S. Mace, F. Kee, et al.
In-Depth Haplotype Analysis of ABCA1 Gene Polymorphisms in Relation to Plasma ApoA1 Levels and Myocardial Infarction
Arterioscler Thromb Vasc Biol, April 1, 2004; 24(4): 775 - 781.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
C.C. Shoulders, E.L. Jones, and R.P. Naoumova
Genetics of familial combined hyperlipidemia and risk of coronary heart disease
Hum. Mol. Genet., April 1, 2004; 13(90001): R149 - 160.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
M. D Tobin, P. S Braund, P. R Burton, J. R Thompson, R. Steeds, K. Channer, S. Cheng, K. Lindpaintner, and N. J Samani
Genotypes and haplotypes predisposing to myocardial infarction: a multilocus case-control study
Eur. Heart J., March 2, 2004; 25(6): 459 - 467.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
D. J Freeman, N. J Samani, V. Wilson, A. D McMahon, P. S Braund, S. Cheng, M. J Caslake, C. J Packard, D. Gaffney, and on behalf of the West of Scotland Study Group
A polymorphism of the cholesteryl ester transfer protein gene predicts cardiovascular events in non-smokers in the West of Scotland Coronary Prevention Study
Eur. Heart J., October 2, 2003; 24(20): 1833 - 1842.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
S. M. Boekholdt and J. F. Thompson
Natural genetic variation as a tool in understanding the role of CETP in lipid levels and disease
J. Lipid Res., July 1, 2003; 44(6): 1080 - 1093.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (54)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Klerkx, A. H.E.M.
Right arrow Articles by Kuivenhoven, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Klerkx, A. H.E.M.
Right arrow Articles by Kuivenhoven, J. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?