Frequency and function of CETP variants among individuals of Asian ancestry

Atherosclerosis 202 (2009) 241–247

Frequency and function of CETP variants among individuals of Asian ancestry John F. Thompson ∗ , Jennifer M. Reynolds, Suzanne P. Williams, Linda S. Wood, Sara A. Paciga, David B. Lloyd Molecular Medicine Department, Pfizer Global Research and Development, Groton, CT 06340, USA Received 11 September 2007; received in revised form 11 March 2008; accepted 12 March 2008 Available online 22 March 2008

Abstract Genetic variation in CETP (cholesteryl ester transfer protein) has been clearly associated with HDL cholesterol levels but its association with cardiovascular disease and related phenotypes has been more controversial, possibly due to variability of polymorphisms and their frequencies across different ethnic populations. To see if there are undetected polymorphisms affecting protein sequence in individuals of Asian ancestry and to determine the functionality of such variants, all exons and adjacent intronic segments were resequenced in 96 individuals and the observed variants cloned and analyzed. Two novel SNPs, including one coding change, S332 to Y332, were identified. Y332 and all other reported variants in Asians were cloned for study in vitro. Secretion efficiency was determined by Western blotting of protein from cell lysates and media. Cholesteryl ester transfer activity was measured in vitro by following the extent of transfer of fluorescently labeled substrate. Y332, Q296 and G442 are all secreted less well than wild type protein but retain significant transfer activity. P151 is not secreted and no transfer activity was detected. These protein variants should all contribute to higher HDL cholesterol in individuals carrying them. Additionally, a splicing variation that causes a protein truncation and non-functional CETP that has been reported predominantly in Asians was also found in two individuals of European ancestry and was on the same haplotype background in the two populations, suggesting a common origin of this null variant. This improved understanding of CETP variation in Asians will allow a more effective comparison of studies across populations. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Pharmacogenetics; CETP; HDL; Association study

1. Introduction The sequence of CETP (cholesteryl ester transfer protein) has long been known to be highly variable. SNPs within both the promoter region and the coding sequence have been shown to have functional effects on CETP activity and expression [1,2]. Some of the most extreme effects have been noted in individuals from Japan and other parts of Asia. The most dramatic example of this may occur in the Omagari region of Japan where over 20% of the population is heterozygous for a splicing mutation (rs5742907) that leads to a truncated, ∗ Corresponding author. Present address: One Kendall Square, Building 700, Helicos Biosciences, Cambridge, MA 02139, USA. Tel.: +1 617 264 1668; fax: +1 617 264 1618. E-mail address: [email protected] (J.F. Thompson).

0021-9150/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2008.03.013

inactive protein [3]. In addition to this splicing variant, there is also an Asian-specific mutation that leads to an amino acid change (D442G, rs2303790) and a protein with altered functional activity [4]. This mutation is also seen throughout different Asian populations with a frequency of greater than 1% in Japan, China, Korea, and Vietnam [5–9]. This SNP has not been seen in populations of European, African, or Indian ancestry [2,10]. There are also literature reports for three other non-synonymous variations (L151P, R282C and L296Q) in Japanese or Chinese individuals that have been linked with HDL cholesterol levels [11,12]. Other promoter and intronic SNPs have been identified but no other coding variants specific for Asians. We have previously sequenced exons from individuals of European and African ancestry so sought to determine if additional Asian-specific coding SNPs could be found and to characterize the properties of all

242

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

proteins generated by these non-synonymous SNPs. Additionally, the frequency of these variations was determined in individuals of European and African ancestry, populations we have resequenced to a similar degree previously [13].

2. Materials and methods 2.1. Samples and genotyping DNA was obtained from multiple sources. The Atorvastatin Comparative Cholesterol Efficacy and Safety Study (ACCESS) was designed to determine the safety and efficacy profile of atorvastatin as compared to other HMG-CoA reductase inhibitors when used to treat patients to NCEP LDL-C criteria [14]. The Treating to New Targets (TNT) study was designed to determine the effect of intensive lipid lowering on patients with stable coronary disease [15]. Whole blood from participating subjects was obtained with appropriate institutional review and appropriate informed consent documentation that defined the study design and provided an assessment of the risks and benefits associated with study participation. DNA from additional individuals of Japanese and Chinese ancestry was obtained from the Coriell Institute for Medical Research (panels HAPMAPPT02, HAPMAPPT05, and HD100CHI). Additional sets of DNA were obtained from Genomics Collaborative Inc. as described previously [2]. For ACCESS and TNT, genomic DNA was extracted from whole blood using the PureGene DNA isolation system (Gentra) per manufacturer’s protocol. Each exon along with short segments of flanking intronic sequence (less than 100 bp) was amplified for sequencing using standard Sanger sequencing techniques with coverage on both strands on all exons and coverage on one or two strands in intronic regions. Novel variants were confirmed by TaqMan genotyping. Some variations discussed herein were first reported elsewhere [2,13,16,17] and were genotyped as described in those publications. Variations reported for the first time here were genotyped using TaqMan technology according to manufacturer’s instructions (Applied Biosystems, Foster City, CA). 2.2. In vitro characterization of variants DNA vectors to express His-tagged versions of CETP with 296Q and 332Y were constructed by overlapping PCR as described previously [16] using primers (5 CGAGTTCAAGGCAGTGCAGGAGACCTGGGGCTTC3 and 5 -GAAGCCCCAGGTCTCCTGCACTGCCTTGAACTCG-3 ) for 296Q, and (5 -CAAGATGCCCAAGATCTACTGCCAAAACAAGG-3 and 5 -CCTTGTTTTGGCAGTAGATCTTGGGCATCTTG-3 ) for 332Y. Anchor primers flanking the unique SanD I and Afl II restriction sites in the CETP cDNA were used to facilitate directional cloning. Transfections into HEK293 cells, generation of cell lysates and media, and partial purification of proteins were carried out as described [16]. Transfected cells were

harvested in physiologically buffered saline (PBS) and whole cell lysates prepared in lysis buffer (50 mM Tris–HCl (pH 8.0), 150 mM NaCl, 24 mM MgCl2 , 0.5 mM Na-o-vanadate, 10% glycerol, 0.1% NP-40). Protein concentrations were estimated using a Bradford assay. Ten micrograms of whole cell lysate and 10 ␮l of concentrated serum-free media were electrophoresed on a 4–12% bis-acrylamide NuPage gel (Invitrogen) and transferred to nitrocellulose using a semidry transfer cell (BioRad). Membranes were probed with a mouse anti-V5 primary antibody (Invitrogen) followed by a goat anti-mouse peroxidase conjugated secondary antibody, and developed using Super Signal West Pico chemiluminescent substrate (Pierce). Blots were analyzed and quantitated on a GeneGnome chemiluminescence detection system (Syngene). Serial dilution of samples and varying detection times were used to assure linearity of response. The activity of each CETP isoform was assessed with a fluorescently tagged (BODIPY-CE) cholesteryl ester transfer assay in vitro [16]. Briefly, bodipy-cholesteryl ester (Molecular Probes) substrate was generated in particles with apolipoprotein A-1 (Biodesign). Ten microliters of concentrated media was combined with HDL (Biomedical Technologies Inc., 80 ␮l of a 160-fold dilution in PBS) in a black 96-well plate, and then incubated for 5 min at 37 ◦ C. After addition of BODIPY-CE (10 ␮l of a four-fold dilution in PBS), the plate was read in a PerkinElmer Victor plate reader (excitation 485 nm, emission 520 nm), then incubated at 37 ◦ C and read every 5 min. Baseline fluorescent values were subtracted from subsequent values.

3. Results DNA from 96 individuals who had participated in either the ACCESS or TNT trial [14,15] was resequenced on both strands across all CETP exons with short segments of flanking introns. Both trials were statin efficacy studies in individuals with or at high risk for cardiovascular disease. All individuals were self-declared East Asian ancestry and unrelated to each other. As listed in Table 1, two novel variations were identified as well as 14 that had been previously identified. Among the novel variations, one results in a non-synonymous change to the protein sequence, S332Y while the other was located in an intron, over 70 bp from a splice site. To determine the frequency of the novel coding variant as well as to better characterize previously identified nsSNPs, all variants were genotyped in Coriell populations of Japanese (n = 91) and Chinese (n = 189) ancestry. Additionally, individuals of European or African American ancestry were also genotyped for comparison with the Asian samples. Some of these CETP variants had been characterized with respect to frequency by us previously [2,13,16,17]. All amino acid changing variants not previously genotyped were assayed using TaqMan with primers and probes listed in Supplementary Table 1 and the resultant frequencies from those assays shown in Table 2. The frequency of promoter and

Yes

Yes Yes Yes

Yes

Yes

rs36122917 rs34855278 rs5880 rs5882 rs5742907 rs2228667 rs2303790 rs1800777 rs5887

rs5881

rs34119551 rs34065661 rs34716057

Allele frequencies of non-synonymous variations in CETP: all previously reported variations that result in a non-synonymous change in amino acid are listed. In the first column, the base change and the position within the exon is provided. If a dbSNP number is available, it is shown in column 2. The actual amino acid change, using the one letter code is listed in column 3. The minor allele frequencies among 91 healthy Japanese, 189 healthy Chinese, and 96 East Asians with CHD from drug trials are listed in the next three columns followed by a summation of that data for “All Asians” in the next column. Frequencies for individuals of European and African American ancestry are provided in the next two columns. If the variation could have arisen by mutation of a CpG dinucleotide, it is noted in the final column.

Asians with CHD Chinese

Table 2 Allele frequency of non-synonymous SNPs in CETP

intronic SNPs such as the commonly studied TaqIB and -629 SNPs and their association with HDL-C has been reported elsewhere for the Asian individuals from the ACCESS study [2]. All variants were genotyped in the ACCESS population (172 individuals of African American ancestry and 2492 of European ancestry) and a subset of SNPs was also genotyped in other populations described in previous publications [2,17]. Of all the non-synonymous variants, the only one with a minor allele frequency of greater than 10% among Asians is I405V (rs5882), which is also the only high frequency nsSNP observed in populations of either European or African ancestry. The next most common SNP, at 1–5%, among Asians and Europeans is A373P (rs5880). Among Africans, an African-selective SNP, R137W (rs34716057), is more common than A373P. Additionally, there is another common nsSNP among African Americans, A-3G (rs34065661), that causes an amino acid change in the signal peptide that is subsequently cleaved upon secretion of CETP so should not affect its specific activity though it could potentially have an effect on secretion efficiency. R451Q (rs1800777) is the only other nsSNP found at greater than 1% among Europeans and is also found in Asians and Africans though at lower frequency. Several other SNPs occur at a frequency of over 0.1% among the different populations. Two variants (L151P, L296Q) identified by others were not detected in our populations. Three variants (R282C, S332Y, and V469M/rs5887) were detected in only a single individual among the thousands genotyped. Because of the rare nature of some of these SNPs, association studies cannot provide conclusive information about the functional impact of these mutations, but in vitro studies can be done to address the question indirectly. In addition to the

All Asians

Variants identified by sequencing: all variants identified among the 96 individuals of Asian ancestry are listed in the first column and whether novel or in the dbSNP database (with an rs number). The base change observed is in column 2 and the minor allele frequency in column 3. The location in the CETP gene and, if coding, the resulting amino acid change are in the final columns.

T + 44A/Ex1 C + 71G/Ex1 C + 21T/Ex5 T + 64C/Ex5 C + 145T/Ex9 T + 8A/Ex10 G + 10A/Ex11 C + 65A/Ex11 A + 152G/Ex11 G + 7A/Ex12 G + 22C/Ex12 A + 16G/Ex14 G + 1A/In14 G + 42A/Ex15 A + 55G/Ex15 G + 82A/Ex15 G + 49A/Ex16

V405I D442G

Yes

CpG

Val > Ile Asp > Gly

European

F270F S332Y A373P

3837 0.02 0.12 0.04 0 0 0 0.02 0 0 0.2 4.4 32.3 0.02 0.02 0 2.9 0.03

Phe > Phe Ser > Tyr Ala > Pro

African American

G161G

455 0.4 7.6 2.3 0 0.3 0 0 0 0.1 0.4 0.8 61.3 0 0 0 0.5 0.4

Gly > Gly

376 0 0 0 0 0 0 0.5 0.1 0 0 1.9 44.4 0.3 0.1 2.4 0.1 0

Intron 5 Exon 6 Intron 6 Intron 7 Intron 7 Intron 8 Exon 9 Exon 11 Exon 12 Intron 12 Exon 14 Exon 15 Intron 15 3 UTR 3 UTR 3 UTR

96 0 0 0 0 0 0 0 0.5 0 0 5.2 40.1 0 0 1 0 0

37.0 0.5 0.5 41.5 1.0 3.6 2.6 0.5 5.2 25.3 33.5 1.0 36.6 17.7 7.5 66.7

Japanese

AG AG CT CT AG CT CT AC CG CT GA AG AG AG CG AG

Protein

91 0 0 0 0 0 0 0 0 0 0 0 47.2 2.2 0.5 3.8 0 0

rs7205804 rs34611098 rs34523084 rs1532625 Novel rs9930761 rs5883 Novel rs5880 rs1800774 rs5882 rs2303790 rs289741 rs1801706 rs289742 rs289743

Coding

Change

Location

N V-12D A-3G R137W L151P R282C L296Q G314S S332Y Y361C V368M A373P I405V COOH V438M D442G R451Q V469M

Frequency

dbSNP

Change

Location

Name

243

189 0 0 0 0 0 0 1.1 0 0 0 1.1 45.2 0 0 2.4 0.3 0

Table 1 Variations identified by sequencing

Yes

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

244

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

Fig. 1. Western analysis of CETP expression. Equivalent amounts of cell lysate and concentrated media were loaded onto 4–12% polyacrylamide gels followed by electrophoresis and visualization using His-tag antibodies. Controls loaded onto the gels included whole cell lysates and concentrated media from cells that had been mock transfected or transfected with empty vector. Two separate isolates of P151 and Y332 were compared to single isolates of wild type CETP, G442, and Q296. Molecular weight markers are present in the left lane.

variants we detected, all known amino acid variants detected by others were tested as comparators. To determine the effect of amino acid variation on secretion and activity, cDNAs for all isoforms and wild type CETP, each with a His-tag, were constructed and transiently transfected into HEK293 cells. Because of the expectation of poor secretion for some of the proteins, both cell lysates and the media were collected for analysis. As shown by Western analysis in Fig. 1, all proteins are expressed at similar levels in cell lysates. In contrast, very different levels of protein are observed secreted into the media. For all secreted proteins, a doublet of bands is observed due to differential glycosylation. Wild type CETP is secreted at the highest levels while P151 is not observed at all in the media. 442G and 296Q can be detected, but at levels at least 10-fold lower than wild type CETP. 332Y is not as well secreted as wild type but significantly better than the other mutated proteins. When these partially purified proteins from the media are assayed for lipid transfer activity using an artificial, fluorescently labeled cholesteryl ester as substrate, wild type protein is more active than all the other forms of the protein. Attempts were made to use equivalent amounts of protein in each assay based on the Western analysis, but the lower yields of the mutated proteins leads to difficulties in generating preparations of similar purity. With that caveat, 332Y is about two-fold less active than wild type and 442G and 296Q about four-fold less active than wild type. No activity could be detected with 151P, whether from concentrated media or cell lysates. In addition to the Asian-specific D442G SNP (rs2303790) described above, another variation (rs5742907, Int14G/A)

has been reported at higher frequency among Asians than among other ethnicities. This mutation eliminates the splice junction site in intron 14, leading to a protein with a C-terminal truncation [18]. Frequencies are generally found to be 0.5–1% in Japanese populations though this can vary (Table 3) and it has been found only sporadically among Europeans. We identified two individuals from separate clinical trials who reported European ancestry and were found to carry a single copy of this mutation. One individual, a 62-year-old female, had a screening HDL-C level of 79 mg/dl, higher than the average (55.2 ± 13.1 mg/dl) among Caucasian females in that population. Additional genotyping of the same individual for several dozen SNPs is also consistent with primarily European rather than Asian ancestry (data not shown). The other individual was from an elderly cohort (minimum age for inclusion was 85) with no lipid data available. To determine whether this rare variation had occurred independently in the European and Asian population groups or was likely to have arisen only once but remained rare, the haplotype for these individuals was determined. DNA from the four individuals to which we had access, two Asians and two Europeans, was cloned as individual alleles and resequenced in that region. Genotypes for the nearby SNPs rs1800774 and rs5882, and the 12 base pair insertion/deletion in intron 12 were assessed. With both European and Asian individuals, the most common haplotypes for rs1800774/rs5882/12bp ins/del are A/C/deletion and A/T/deletion, each with about 30% frequency. Most of the remaining individuals have G/C/deletion or G/C/insertion. For both Japanese and European individuals, the rare allele

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

245

Table 3 Minor allele frequency for D442G and Int14G/A SNPs in multiple populations

Minor allele frequency of D442G (rs2303790) and G + 1A/In14 (rs5742907): studies in which either D442G or G+1/In14 were genotyped are listed by country or origin and in decreasing order of the number of individuals genotyped. Studies are included only if they were not family-based and the individuals were not directly ascertained using HDL for inclusion/exclusion. Selection of patients based on cardiovascular disease status may alter the distribution of CETP genotypes but that effect would be not be as pronounced as if HDL were used directly. In some references, per cent heterozygosity was used rather than minor allele frequency. If heterozygosity was originally reported, it was transformed to minor allele frequency assuming perfect adherence to Hardy–Weinberg principles. If no value is given, the study did not genotype that SNP. The number of individuals genotyped in a given group is provided in the final column. When cases and controls were listed separately in the original reference, they are provided as a separate entry in the table. Studies are arranged by size of population studied with sets of studies from the same region shaded as blocks. Literature reference: [5], [6], [3], [23], [24], [27], [29], [30], [32], [29], [3], [7], [25], [26], [28], [33], [34], [34], [35], [28], [26], [36], [26], [8], [31], [9], [10], [36], [22].

for rs5742907 occurs on a background of A/C/deletion. This is consistent with, though does not prove, a common origin for the SNP rather than independent occurrences.

4. Discussion The high level of variability within CETP has made it one of the most highly studied genes for association with clinical phenotypes. The functional promoter polymorphisms, altered splice sites, and amino acid changing variations have provided a wealth of data. Unfortunately, not all studies have yielded consistent results. Often, it is simply a matter of insufficient population size, but discrepant results may also be caused by differences in individual ethnicity or the inclusion/exclusion criteria for particular studies. Many CETP SNPs have not been well characterized across multiple ethnicities so assessing how this might impact association

studies has not been possible. By determining which SNPs are truly ethnic-specific, what their functional impact is likely to be, and how frequently they are likely to occur, it is possible to better account for differences in populations across studies. The low frequency of the rare variants precludes their inclusion in LD blocks as normally defined and also makes assessment of Hardy–Weinberg equilibrium less meaningful. Additionally, the newly characterized variants are too infrequent to carry out meaningful association studies with clinical phenotypes including HDL-C. Even combining the novel rare variations into a single pool would not provide a large enough group for statistical significance and would be further complicated by the very different ascertainment criteria that were used across populations. Thus, in vitro characterization is necessary to understand their functional characteristics. As shown in Table 2, the only nsSNPs that occur at over 1% frequency among Asians are A373P (rs5880), I405V

246

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

Fig. 2. Transfer of fluorescent cholesteryl ester by CETP variants. The ability of equivalent amounts of partially purified CETP isoforms to transfer fluorescently labeled cholesteryl ester from a high concentration, quenched particle to a particle with no label and thus no quenching was assessed for the same variants shown in Fig. 1. Measurements were made every 5 min with saturation of transfer occurring beyond 1 h.

(rs5882), and D442G (rs2303790). The first two are also found at relatively high frequencies among individuals of African and European ancestry. CETP/442G (rs2303790) has been found in all East Asian populations studied and is less active than the more common CETP/442D [4]. CETP/151P was previously found in one Japanese individual [11] but was not detected by us. CETP/296Q was previously found in Chinese individuals [12] but not detected by us. Its activity in vitro has not been reported. Of the SNPs with a frequency of greater than 1%, only D442G appears to be unique among Asians. The association of CETP/442G with HDL cholesterol levels has been long established and the cause for this is likely to be its poor secretion and thus lower levels in the plasma. In addition to the secretion defect, we have also characterized it in vitro and found a lower specific activity for this isoform relative to wild type (Fig. 2). The association of 442G with disease or longevity has been less clear with some reports finding an association with higher levels of disease while others have not [1]. However, the largest, prospective study examining this SNP (in 2340 elderly Japanese Americans) found those with this SNP had a lower level of CHD than did those with the more common allele [19]. Comparison across studies for D442G may be particularly problematic due to geographical frequency variations. As shown in Table 3, the frequency of D442G has been found to be highly variable, leading to potential problems with matching of cases and controls. Table 3 provides a summary of literature studies in which individuals were assessed for D442G. Studies that were family-based and studies in which the individuals were initially ascertained based on HDL-C levels are omitted. Either of these factors could skew the frequency observations. Among the Japanese populations, D442G minor allele frequency varies from 1.5 to 6.9%. Among Chinese populations, the variation is similar, ranging from 1.5 to 5.4%. Determination of the relative frequency

of D442G in Japanese compared to Chinese will require additional, equivalently ascertained, population-based studies. Koreans may have the highest rate of D442G as the two studies thus far reported minor allele frequencies of 4.6 and 5.6%, but these were relatively small studies. Some of this frequency variation is due to small sample size, but a geographical component also seems to be present. Even though many thousands of individuals have been examined, this SNP has not been found among populations of non-Asian ancestry. Another variation that is more common among Asians is rs5742907 (G + 1A/In14) that occurs at a splice site and thus significantly alters the C-terminal sequence of CETP. This SNP appears to be even more geographically variable than D442G. The largest population-based studies among Japanese place the frequency at 0.25% minor allele frequency [6] but other studies of Japanese individuals have found minor allele frequencies ranging from 0 to 13.5%. This variation has been less well studied but has also been reported among Chinese individuals and the SNP has been observed sporadically among individuals of European ancestry [20]. We identified two individuals of European ancestry from two different studies who were heterozygous for this mutation. This particular mutation occurs at a CpG site (Table 2), a sequence previously shown to be much more highly mutable than other dinucleotide sequences [21], consistent with the possibility of an independent origin of the mutation. However, all four individuals examined with the rare allele were found to have it on the same haplotype background. This haplotype occurs at a frequency of about 30% in both ethnicities so additional work would be necessary to confirm the question of common ancestry. In the populations we studied, there are no other Asianspecific SNPs that occur at a frequency of higher than 1%. L296Q was previously reported to be common among one group of Chinese individuals [12], but this does not appear to be true of all geographical regions of China as we were unable to detect it in our samples. It may have a locally high frequency like rs5742907 in Omagari [3] but it cannot be common throughout China. The high level of genetic variation in both the coding sequence and the promoter regions of CETP provides a valuable tool for understanding its function as long as care is taken to match both the geographical origin and disease status of individuals being compared. The CETP variants studied here and elsewhere show that different versions of the protein can have altered secretion properties as well as altered activity and thus can impact many clinical phenotypes. Having an understanding of the frequency and function of these variations across populations provides a means of comparing studies.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.atherosclerosis. 2008.03.013.

J.F. Thompson et al. / Atherosclerosis 202 (2009) 241–247

References [1] Boekholdt SM, Thompson JF. Natural genetic variation as a tool in understanding the role of CETP in lipid levels and disease. J Lipid Res 2003;44:1080–93. [2] Thompson JF, Wood LS, Pickering EH, DeChairo B, Hyde CL. High density genotyping and functional SNP localization in the CETP gene. J Lipid Res 2007;48:434–43. [3] Hirano K, Yamashita S, Nakajima N, et al. Genetic cholesteryl ester transfer protein deficiency is extremely frequent in the Omagari area of Japan. Marked hyperalphalipoproteinemia caused by CETP gene mutation is not associated with longevity. Arterioscler Thromb Vasc Biol 1997;17:1053–9. [4] Takahashi K, Jiang XC, Sakai N, et al. A missense mutation in the cholesteryl ester transfer protein gene with possible dominant effects on plasma high density lipoproteins. J Clin Invest 1993;92:2060–4. [5] Zhong S, Sharp DS, Grove JS, et al. Increased coronary heart disease in Japanese-American men with mutation in the cholesteryl ester transfer protein gene despite increased HDL levels. J Clin Invest 1996;97:2917–23. [6] Arai H, Yamamoto A, Matsuzawa Y, et al. Polymorphisms in four genes related to triglyceride and HDL-cholesterol levels in the general Japanese population in 2000. J Atheroscler Thromb 2005;12:240–50. [7] Hsu L-A, Ko Y-L, Hsu K-H, Ko Y-H, Lee Y-S. Genetic variations in the cholesteryl ester transfer protein gene and high density lipoprotein cholesterol in Taiwanese Chinese. Hum Genet 2002;110:57–63. [8] Song GJ, Han GH, Chae JJ, et al. The effects of the cholesterol ester transfer protein gene and environmental factors on the plasma high density lipoprotein cholesterol levels in the Korean population. Mol Cells 1997;7:615–9. [9] Thu NN, Mai TT, Ohmori R, et al. Effect of the cholesteryl ester transfer protein genotypes on plasma lipid and lipoprotein levels in Vietnamese children. Pediatr Res 2005;58:1249–53. [10] Dixit M, Mittal B. Frequencies of CETP gene TaqI B and D442G polymorphisms in North Indian population. Curr Sci 2005;88:1973–6. [11] Nagano M, Yamashita S, Hirano K, et al. Two novel missense mutations in the CETP gene in Japanese hyperalphaproteinemic subjects: high throughput assay by Invader assay. J Lipid Res 2002;43:1011–8. [12] Zheng K, Zhang S, Zhang L, et al. Carriers of three polymorphisms of cholesteryl ester transfer protein gene are at increased risk to coronary heart disease in a Chinese population. Int J Cardiol 2005;103:259–65. [13] Lloyd DB, Reynolds JM, Cronan MT, et al. Novel variants in human and monkey CETP. Biochim Biophys Acta 2005;1737:69–75. [14] Ballantyne CM, Andrews TC, Hsia JA, Kramer JH, Shear C. Correlation of non-high-density lipoprotein cholesterol with apolipoprotein B: effect of 5 hydroxymethylglutaryl coenzyme A reductase inhibitors on non-high-density lipoprotein cholesterol levels. Am J Cardiol 2001;88:265–9. [15] LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med 2005;352:1425–35. [16] Lloyd DB, Lira ME, Wood LS, et al. Cholesteryl ester transfer protein variants have differential stability but uniform inhibition by torcetrapib. J Biol Chem 2005;280:14918–22. [17] Thompson JF, Durham LK, Lira ME, Shear C, Milos PM. CETP polymorphisms associated with HDL cholesterol may differ from those associated with cardiovascular disease. Atherosclerosis 2005;181:45–53. [18] Brown ML, Inazu A, Hesler CB, et al. Molecular basis of lipid transfer deficiency in a family with increased high-density lipoproteins. Nature 1989;342:448–51.

247

[19] Curb JD, Abbott RD, Rodriguez BL, et al. A prospective study of HDLC and cholesteryl ester transfer protein gene mutations and the risk of coronary heart disease in the elderly. J Lipid Res 2004;45:948–53. [20] Fumeron F, Betoulle D, Luc G, et al. 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 1995;96:1664–71. [21] Hacia JG, Fan J-B, Ryder O, et al. Determination of ancestral alleles for human single-nucleotide polymorphisms using high-density arrays. Nat Genet 1999;22:164–7. [22] Freeman DJ, Samani NJ, Wilson V, et al. 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 2003;24:1833–42. [23] Arashiro R, Katsuren K, Maung KK, Fukuyama S, Ohta T. Effect of a common mutation (D442G) of the cholesteryl ester transfer protein gene on lipids and lipoproteins in children. Pediatr Res 2001;50:455–9. [24] Kimura H, Gejyo F, Suzuki S, et al. Cholesteryl ester transfer protein as a protective factor against vascular disease in hemodialysis patients. Kidney Int 1999;56:S186–9. [25] Hui S-P. Frequency and effect on plasma lipoprotein metabolism of a mutation in the cholesteryl ester transfer protein gene in the Chinese. Hokkaido J Med Sci 1997;72:319–27. [26] Zhuang Y, Wang J, Qiang H, et al. Cholesteryl ester transfer protein levels and gene deficiency in Chinese patients with cardio-cerebrovascular diseases. Chin Med J 2002;115:371–4. [27] Haraki T, Inazu A, Yagi K, Kajinami K, Koizumi J, Mabuchi H. Clinical characteristics of double heterozygotes with familial hypercholesterolemia and cholesteryl ester transfer protein deficiency. Atherosclerosis 1997;132:229–36. [28] Wu JH, Lee Y-T, Hsu H-C, Hsieh L-L. Influence of CETP gene variation on plasma lipid levels and coronary heart disease: a survey in Taiwan. Atherosclerosis 2001;159:451–8. [29] Arai Y, Hirose N, Yamamura K, et al. Deficiency of cholesteryl ester transfer protein and gene polymorphisms of lipoprotein lipase and hepatic lipase are not associated with longevity. J Mol Med 2003;81:102–9. [30] Inazu A, Jiang X-C, Haraki T, et al. 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 1994;94:1872–82. [31] Choi HS, Park JB, Han KO, et al. A common mutation in cholesteryl ester transfer protein gene and plasma HDL cholesterol level before and after hormone replacement therapy in Korean postmenopausal women. Korean J Int Med 2002;17:83–7. [32] Akita H, Chiba H, Tsuchihashi K, et al. Cholesteryl ester transfer protein gene: two common mutations and their effect on plasma high-density lipoprotein cholesterol content. J Clin Endo Metab 1994;79:1615–8. [33] Jap T-S, Wu Y-C, Tso Y-C, Chiu C-Y. A novel mutation in the intron 1 splice donor site of the cholesterol ester transfer protein (CETP) gene as a cause of hyperalphalipoproteinemia. Metabolism 2002;51:394–7. [34] Zheng K-Q, Zhang S-Z, Zhang L, Huang D-J, Liao L-C, Hou Y-P. A novel missense mutation (L296Q) in cholesteryl ester transfer protein gene related to coronary heart disease. Acta Biochim Biophys Sin 2004;36:33–6. [35] Wang J, Qiang H, Chen D, Zhang C, Zhuang Y. CETP gene mutation (D442G) increases low-density lipoprotein particle size in patients with coronary heart disease. Clin Chim Acta 2002;322:85–90. [36] Xia Y, Baum L, Pang CP, Siest G, Visvikis S. Cardiovascular riskassociated allele frequencies for 15 genes in healthy French and Chinese. Clin Chem Lab Med 2005;43:817–22.