点击显示 收起
【摘要】
Objective- To identify common variations in genes in the reverse cholesterol transport pathway with nongender-specific influence on plasma lipid and apolipoprotein levels.
Methods and Results- An average of 5 single nucleotide polymorphisms (SNPs) were genotyped within each of 45 genomic regions (54 genes) in blacks (1131 females and 812 males) and whites (1102 females and 954 males) from the Coronary Artery Risk Development in Young Adults (CARDIA) study. SNPs and gene-based 3-SNP haplotypes were evaluated for their ability to predict variation in plasma apolipoproteins (apo) A-I and apoB, total cholesterol (TC), high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglycerides (TG). We identified 14 SNPs in 6 candidate gene regions that explained statistically significant variation in the same trait in both genders of at least one race and with evidence of consistent genotype mean trend across gender within race. Haplotype analyses identified 9 candidate gene regions that explained statistically significant variation in one or both races.
Conclusion- Four gene regions, ABCA1, APOA1/C3/A4/A5, APOE/C1/C4/C2, and CETP, explained plasma lipoprotein variation most consistently across strata. Other gene regions that influence plasma lipid and apolipoprotein levels within race include CYP7A1, LPL, PPARA, SOAT1, and SREBF2.
To identify common gene variations with a consistent influence we evaluated 262 SNPs in 54 genes for association with 6 plasma lipid and apolipoprotein traits. We identified 14 SNPs and 9 gene regions with genotype effects.
【关键词】 lipids genetics of cardiovascular disease lipid and lipoprotein metabolism
Introduction
Plasma measures of lipid metabolism are well-established risk factors of coronary artery disease (CAD), the leading cause of death worldwide. 1 Increased risk of CAD has been associated with elevated triglycerides (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C), and decreased high-density lipoprotein cholesterol (HDL-C). 2-6 CAD has also been associated with low plasma apolipoprotein (apo) A-I and high apoB. 7,8 These lipids and apolipoproteins are regulated by complex metabolic and cellular processes, which, in turn, are influenced by both genetic and environmental factors. 9,10
Reverse cholesterol transport (RCT) is the process by which cholesterol is effluxed from peripheral tissues onto acceptor particles, primarily HDL, in the plasma for uptake by the liver. RCT involves numerous lipid transfer proteins, enzymes, apolipoproteins, and membrane-bound receptors. 11 The genes encoding these proteins, as well as genes encoding proteins that regulate their transcription, are candidates for influencing variation in plasma levels of apoA-I, apoB, HDL-C, LDL-C, TC, and TG. Based on a model RCT pathway, 12,13 we identified a set of 54 genes involved in RCT for evaluating the impact of genetic variation on variation in plasma lipid and lipoprotein levels.
Genetic association studies have been plagued by inconsistent results. Two sources of these inconsistencies are false-positives (ie, type I error) and context-dependency of genotype effects. 14-16 In this study, we sought to identify gene regions having nongender-specific genotype effects on plasma lipid and apolipoprotein levels within population-based samples of black and white young adults (aged 18 to 30).
Methods
Sample and Laboratory Measurements
Details of the Coronary Artery Risk Development in Young Adults (CARDIA) study have been described elsewhere. 17,18 In brief, young adults aged 18 to 30 years were randomly sampled from the total community in Birmingham, Ala; from selected census tracts in Chicago, Ill, and Minneapolis, Minn; and from the Kaiser-Permanente health plan membership in Oakland, Calif. Participants were recruited to be roughly equal in racial, gender, age ( 24 years) and education (high school or less versus higher than high school) groups. Study participants were given 6 sequential examinations from the time of the study initiation (1985 to 1986). Results shown here pertain to data collected at the first examination. All participants gave written informed consent, including consent for genetic studies. The CARDIA study was approved by the Institutional Review Boards of the 4 participating field centers, and this ancillary study was approved by additional Institutional Review Boards.
Venous blood was drawn after a 12-hour fast. TC and TG were measured enzymatically. 19 HDL-C was determined by precipitation with dextran sulfate/magnesium chloride. 20 LDL-C was calculated using the Friedewald equation, after excluding individuals whose plasma triglycerides exceeded 400 mg/dL. 21 Body mass index was calculated as weight (kg) divided by height squared (m 2 ). apoA-I and apoB were determined using radioimmunoassay. 22
Single Nucleotide Polymorphisms Selection and Genotyping
Single nucleotide polymorphisms (SNPs) were identified from NCBI dbSNP and Celera Discovery System databases in 54 genes belonging to a pathway model of RCT. 13 An average of 5 SNPs per gene were selected for genotyping by identifying SNPs spaced evenly across the gene with relative frequencies of the rare allele reported in the dbSNP database (http://www.ncbi.nlm.nih.gov/SNP/) to be 5% for whites or blacks. Of 272 SNPs selected by this method and genotyped as below, 262 (96.4%) were retained after general quality control measures were applied (missing data 98.5% concordance in blind duplicate analysis). Average spacing between SNPs within a gene was 9.06 Kb (range, 41 bp to 90.64 Kb).
For the present study, genotypes were obtained on 3993 individuals (1132 black females, 807 black males, 1101 white females, and 953 white males). Genotyping was performed using polymerase chain reaction amplification of genomic DNA, a short extension reaction across the polymorphic site, and mass spectrometry to detect allele-specific mass differences. Allele detection and genotype calling was performed using a MassARRAY System from Sequenom (San Diego, Calif). The sequences of the polymerase chain reaction and extension primers are available from the authors on request.
Statistical Analysis
Allele frequencies were obtained by direct counting. Hardy-Weinberg equilibrium was evaluated using a chi-square goodness of fit test. Individuals were removed from the analyses if they had not fasted 8 hours or more before the examination. TG levels were ln transformed to reduce skewness. Individuals with extreme values (±3 SD) were also removed. All lipid and apolipoprotein values were adjusted before analysis by fitting a race-, field center- and gender-specific linear regression model containing age, age 2, age 3, and body mass index. The residuals from the regression model were added back to the race- and gender-specific grand mean to produce an adjusted phenotype value for each individual. For the haplotype analyses, values were further adjusted for gender effects by adding the residuals from a model containing only gender back to the race-specific grand mean.
Differences in quantitative trait means among genotypes were evaluated by a 1-way ANOVA. 23 Homogeneity of variances among genotypes was evaluated using Levene test. 24 Where variances were unequal, Welch modified F statistic was used in assessing significance. 25 To partially control for type 1 error in the single-SNP tests, a probability value from the 1-way ANOVA 0.01 in both genders of a race, or P 0.001 in one gender with P 0.05 in the other, was required for further consideration of genotype mean trends. Homogeneity of genotype mean trends was evaluated by ANOVA using a model that included genotype, gender, and genotype-by-gender interaction terms. Homogeneity of genotype mean trends across genders was rejected if P 0.05 for the interaction term. Consistency has become the sin quo non for evaluating significance of genetic association studies in light of continuing difficulties with multiple comparisons. In this study, we defined consistency as statistically significant and directionally similar effects in both genders within a race.
Haplotype frequencies within race were inferred from unphased genotype data using the expectation maximization algorithm. 26,27 Haplotype trend regression, as implemented in the Helixtree (Golden Helix, Bozeman, Mont) software package, was used to evaluate association between plasma lipid and apolipoprotein levels and the haplotype probabilities of individuals. 28 In this method, a matrix, D, of the inferred conditional probabilities of haplotypes given genotypes for each individual in the sample is created. A regression equation, E(Y)=Dß, is then used to relate inferred haplotype probabilities to the dependent variable. In this equation, Y is the matrix of individual response values, and D is the matrix of inferred conditional probabilities of haplotypes given the genotype. The conditional probability of a pair of haplotypes (for example h 2,h 3 ) for the i th individual with genotype G i is
where Ph u Ph v denote haplotype frequencies. An F-test was used to test the overall null hypothesis H 0 : ß 1 =ß 2 =... =ß n =0 for n haplotypes in the population. Within gene regions, a 3-SNP haplotype sliding-window method was used to evaluate haplotype effects across the region. Adjustment for multiple testing of raw probability values was performed using the false discovery rate method suggested by Benjamini and Hochberg 29 as implemented in the Helixtree software.
Results
Table 1 summarizes the characteristics of the CARDIA participants genotyped for this study. Males had higher ( P 0.05) plasma levels of TG, whereas females had higher levels of apoA-I and HDL-C. Black females had higher plasma TC than males. White males had higher plasma apoB and LDL-C than females.
TABLE 1. Characteristics (Mean and Standard Deviation) of 1939 Black and 2054 White Individuals of the Coronary Artery Risk Development in Young Adults (CARDIA) Study
Characteristics of the genotyped SNPs are summarized in Table 2. Of the 262 SNPs, 16% were exonic (61% nonsynonymous). Genotype frequencies for 2 SNPs in blacks and 9 SNPs in whites were out of Hardy-Weinberg equilibrium ( P 0.001). In all except 2 cases, this was because of fewer than expected heterozygotes. In no case did results for SNPs not in Hardy-Weinberg equilibrium fit our criteria for consistency of association, nor were they involved in defining a 3-SNP haplotype having a statistically significant phenotypic effect.
TABLE 2. Characteristics of 262 SNPs in 45 Gene Regions
Seventy-four of 262 SNPs were associated ( P 0.05) with quantitative traits in black females, 68 in black males, 61 in white females, and 75 in white males (7% of 6280 valid statistical tests). The same trait was associated in both genders ( P 0.01 in both genders, or P 0.001 in one gender with P 0.05 in the other) with variation in 11 SNPs in blacks and 8 SNPs in whites. These SNPs were evaluated for evidence of homogeneity of mean trends across gender strata using a nominal probability value of 0.05 as the threshold of significance. In blacks, homogeneity of apoB, LDL-C, and TC mean trends across gender was rejected ( P 0.05) for apolipoprotein E/C1/C2/C4 gene cluster ( APOE/C1/C2/C4 ) SNP rs741780 and homogeneity of LDL-C mean trends was rejected for APOE/C1/C2/C4 SNP rs405509.
Tables 3 and 4 summarize the pattern of significant associations between the 6 lipid traits and the 14 SNPs with evidence of a consistent effect across genders. Genotype means for the 14 SNPs in Tables 3 and 4 are presented in online supplemental Table I (see http://atvb.ahajournals.org). For 1 SNPs in 5 gene regions, genotype mean trends were consistent across genders within race for at least one HDL-related plasma trait ( Table 3 ). For 1 SNPs in 2 gene regions, genotype mean trends were consistent across genders within race for at least 1 LDL-related plasma trait ( Table 4 ).
TABLE 3. The P Values From ANOVA Tests of Association for 8 SNPs and 3 HDL-Related Plasma Traits
TABLE 4. P Values From ANOVA Tests of Association for 7 SNPs and 3 LDL-Related Plasma Traits
Results of the gene-based haplotype sliding window analyses are plotted in Figure 1 for the HDL-related traits, plasma apoA-I, HDL-C, and lnTG, and in Figure 2 for the LDL-related traits plasma apoB, LDL-C, and TC. These results are plotted as the -log10 false discovery rate-adjusted probability values from the haplotype trend regression analyses. 3-SNP haplotypes are arranged on the x-axis by the location from the p-terminal of chromosome 1 to the q-terminal of chromosome 21. Haplotypes 150 000 bp apart. For ease of comparison, all gene regions are plotted on the same axis. The threshold for statistical significance, an false discovery rate-adjusted P 0.05, is indicated by a horizontal red line.
Figure 1. Results of haplotype sliding window analyses of the HDL-related traits plasma apoAI, HDL-C, and lnTG related as -log10 false discovery rate (FDR)-adjusted probability values from 3-SNP haplotype trend regression analyses.
Figure 2. Results of haplotype sliding window analyses of the LDL-related traits plasma apoB, LDL-C, and total cholesterol (TC) related as -log10 false discovery rate (FDR)-adjusted probability values from 3-SNP haplotype trend regression analyses.
Nine gene regions were statistically significantly associated with at least 1 plasma lipid or apolipoprotein trait in the haplotype trend regression analyses. In both races, HDL-C was associated with ATP-binding cassette transporter A1 ( ABCA1 ), apolipoprotein A-I/apolipoprotein C-III/apolipoprotein A-IV/apolipoprotein A-V ( APOA1/C3/A4/A5 ) cluster, and cholesteryl ester transfer protein ( CETP ) gene-based haplotypes. Also in both races, plasma apoB, LDL-C, and TC were associated with APOE/C1/C4/C2 cluster haplotypes. Unique to blacks were plasma apoA-I associations with cytochrome p450, family7, subfamily A, polypeptide 1 ( CYP7A1 ), plasma HDL-C with APOE/C1/C4/C2, lipoprotein lipase ( LPL ), and sterol o-acyltransferase 1 ( SOAT1 ), and plasma LDL-C with peroxisome proliferator-activated receptor alpha ( PPARA ) haplotypes. Unique to whites were plasma lnTG associations with APOA1/C3/A4/A5 and a plasma apoB association with sterol regulatory element binding transcription factor 2 ( SREBF2 ) haplotypes.
Discussion
We evaluated the nongender-specific association of variation in 6 plasma lipid and apolipoprotein traits with variation in 262 SNPs in 54 genes involved in RCT. Four gene regions, ABCA1, APOA1/C3/A4/A5, APOE/C1/C4/C2, and CETP, stand out from these analyses for consistency of association across samples and strata. Race-specific effects were observed for CYP7A1, LPL, PPARA, SOAT1, and SREBF2.
ABCA1 transports cellular cholesterol across the cell membrane to plasma acceptor particles such as apoA-I. 30,31 In CARDIA blacks, HDL-C was consistently associated with ABCA1 SNP rs2515602, located at one peak of significance in the haplotype analyses. This region overlaps with the haplotype analysis peak location in whites. Together, significant haplotypes span approximately the 9 th through the 18 th exons of ABCA1, a region with high average LD in both races (average D'=0.90 in blacks and 0.99 in whites). SNPs rs2066718 (V771 mol/L) and rs2065412 were in the best haplotypes of both races. An independent effect of V771 mol/L on HDL-C has been previously reported, 32 but not consistently. 33 In CARDIA, genotype at V771 mol/L does not fully explain the HDL-C mean differences among individuals containing 2, 1, or 0 copies of common haplotypes (not shown).
apoA-I, apoC-III, apoA-IV, and apoA-V are components of plasma HDL and triglyceride-rich lipoproteins. In lipid metabolism, apoA-I serves as a cholesterol acceptor particle, apoC-III is a lipoprotein lipase inhibitor, and apoA-IV and apoA-V have been associated with variation in plasma TG, HDL-C, and TC levels. 34 We previously reported the effect of APOA5 gene variation on plasma lnTG levels in CARDIA, 35 and evaluation of additional SNPs in the region did not alter our conclusions. These analyses indicate that variation in the APOA1/C3/A4/A5 gene region influences HDL-C levels in CARDIA, consistent with previously reported associations of dyslipidemia 36 and HDL-C. 37-40 The 3 haplotypes significant in the sliding window analysis spanned the APOA1 and APOA4 genes in whites, and the APOC3 gene in blacks. SNP coverage was sparse in this region (an average of just 4 SNPs per gene) and additional genotyping will be required to differentiate among genes in the region for effect on HDL-C.
Results from APOE/C1/C4/C2 gene region analyses in this study were in accord with the well-reported combined effect of APOE SNPs rs7412 and rs429358 (the 2/3/4 protein isoforms) on LDL-related plasma traits 41-44 and with APOE/C1/C4/C2 gene region effects on HDL-C. 44.
Cholesteryl ester transfer protein is an important regulator of the movement of cholesteryl esters and triglycerides among lipoprotein particles. 45 In these results, most prominent was the region of CETP containing the rs708272 (Taq1B) polymorphism, previously shown to be in strong LD with at least 1 promoter region polymorphism that may influence CETP concentration. 46,47 In CARDIA, haplotype analyses were consistent with this finding (HDL-C means differed by copy number of common haplotypes containing the A allele of rs708272, P <0.05), but were also consistent with the presence of additional functional polymorphisms (HDL-C means also differed by copy number of some common haplotypes with the G allele at that locus P <0.05). Meta analysis has confirmed more than one CETP region polymorphism with an effect on HDL-C level consistent across samples. 45
The protein encoded by CYP7A1 is the rate-limiting enzyme in bile acid synthesis. 48 In blacks, consistent CYP7A1 associations with LDL-C and TC, characterized by increased levels in carriers of the rare alleles, were in accord with the previous reports of 5' region SNP effects. 49-51 Plasma apoA-I was associated with a 5' region CYP7A1 haplotype (rs1023649, rs8192871, rs11786580) in blacks ( P =0.0001). Tests of equality of apoA-I means in individuals classed as having 0, 1, or 2 copies were significant ( P <0.05) for haplotypes distinguished by an A allele at rs1023649. Although not consistently significant by the criteria applied in this study, association with apoA-I level was nominally significant ( P <0.05) for rs8192871 in European-American males. Associations with plasma apoA-I and with HDL-C have been previously shown for CYP7A1 gene region polymorphisms in Pacific Islanders 52 and in Asians, 53 respectively. These findings support evidence of a CYP7A1 influence on HDL-related traits in addition to the influence on LDL-C and TC.
Lipoprotein lipase hydrolyzes triglycerides in plasma chylomicron and very-low-density lipoprotein particles. 54 LPL gene variation has been reported to influence both plasma HDL-C and TG levels. 55-56 In CARDIA blacks, HDL-C was associated with 5 consecutive 3-SNP combinations spanning the first intron through the ninth codon (involving 6 of 8 measured SNPs). The nominal probability values from the haplotype trend regression analyses of these 5 haplotypes were uniformly lower (nominal P =0.0005 to P =0.0008) than probability values for single-SNP tests in the gender-stratified analyses (nominal P =0.0043 to P =0.8355 in females, and P =0.0086 to P =0.9043 in males). Although not consistently significant, associations between HDL-C and LPL variation were also observed in whites (hCV9642874 P =0.034 in females and P =0.008 in males; rs13266204 P =0.033 in females and P =0.015 in males). SNPs nominally significant in whites are in the first intron, whereas SNPs consistently significant in blacks are nearly 25 kb distal within the eighth intron and the 3' untranslated region. Sequence analysis of the LPL gene in blacks, whites, and Finns indicates a recombination hotspot between these 2 regions. 57,58 Multiple LPL variable sites may function to influence HDL-C variation.
The peroxisome proliferator-activated receptor PPARA regulates fatty acid catabolism in the liver and other tissues. 59 LDL-C was associated with a PPARA haplotype in blacks, but this "haplotype" consisted of a single 3' SNP, rs1053332, for which LDL-C means increased with copy number of the rare allele ( P =0.0013). Coding region polymorphisms in PPARA have been associated with plasma levels of apoB, LDL-C, and TC in the Framingham Offspring Study, 60 but those polymorphisms were not measured in this study. LD (measured as D') between the 3 coding region SNPs measured and rs1053332 averaged 0.52 in blacks. Further investigation will be required to assess whether rs1053332 associations were attributable to LD with previously studied polymorphic sites.
SOAT1 functions intracellularly to convert cholesterol into cholesteryl esters. 61,62 HDL-C was associated in blacks with a haplotype of SOAT1 consisting of rs2783391, rs2247071, and rs2493121, a promoter region and 2 intronic polymorphisms, respectively. A "C" allele at both rs2783391 and rs2247071 distinguished haplotypes with significantly different copy number class means from other common haplotypes. Although not consistently significant, HDL-C was associated with rs2862616 in black males ( P =0.0294), whereas apoA-I level was associated with rs2783391, rs2247071, and rs2255375 in white males ( P =0.0008, P =0.0281, and P =0.0022, respectively). Although LD across the region was not high (average D'=0.68 in blacks and 0.78 in whites), it remains most likely that these associations are attributable to LD with 1 unmeasured polymorphic sites that function to alter gene expression or structure. Outside of this study, little work has been done to characterize the relationship between SOAT1 polymorphism and plasma lipid metabolism in humans.
The SREBF2 transcription factor regulates genes involved in cholesterol homeostasis. 63 In this study, an association between plasma apoB level and a single SREBF2 haplotype (rs1009544, rs1883205, and rs2284083), entirely in the first intron, was unique to whites. Associations with TC and LDL-C have been previously reported for SREBF2 region polymorphisms, 64,65 and these finding are further support of a role for this gene in LDL metabolism.
Although the sparse SNP coverage of many of the genes in this study precludes the ability to rule out any gene as having an effect on plasma lipid metabolism, these results suggest the presence of influential variations in at least 9 gene regions. Despite the complexity of genetic and environmental control of lipid metabolism, some genes are likely to influence these CAD risk factors in a manner that is consistent across genders. A more comprehensive evaluation of potential functional variations in and around these genes is warranted.
Acknowledgments
Sources of Funding
This work was supported by NIH grants HL072905 HL072810, HL072904, and GM065509.
Disclosures
None.
【参考文献】
Am Heart Association. Heart Disease and Stroke Statistics - 2005 update. Dallas, Tex. 2005.
Austin MA, Hokanson JE, Edwards KL. Hypertriglyceridemia as a cardiovascular risk factor. Am J Cardiol. 1998; 81: 7B-12B
Castelli WP, Anderson K, Wilson PW, Levy D. Lipids and risk of coronary heart disease: The Framingham Study. Ann Epidemiol. 1992; 2: 23-28.
Gordon DJ, Rifkind BM. High-density lipoprotein: the clinical implications of recent studies. N England J Med. 1989; 321: 1311-1316.
Davignon J, Cohn JS. Triglycerides: a risk factor for coronary heart disease. Atherosclerosis. 1996; 124: S57-S64.
Miller GJ, Miller NE. Plasma-high-density-lipoprotein concentration and development of ischemic heart disease. Lancet. 1975; 1: 16-19.
Brunzell JD, Sniderman AD, Albers JJ, Kwiterovich PO Jr. apolipoproteins B and A-I and coronary artery disease in humans. Arterioscler. 1984; 4: 79-83.
Kottke BA, Holmes DR, Mao SJT, Kneller RW, Hallaway BJ, Zinsmeister AR. apolipoproteins and coronary heart disease. Circulation. 1985; 71: 415A
Lusis AJ. Athersoclerosis Nature. 2000; 407: 233-237.
Sing CF, Stengard JH, Kardia SL. Genes, environment and coronary artery disease. Atherosler Thromb Vasc Biol. 2003; 23: 1190-1196.
Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995; 36: 211-228.
von Eckardstein A, Nofer J-R, Assmann G. High density lipoproteins and arteriosclerosis: Role of cholesterol efflux and reverse cholesterol transport. Arterioscler Thromb Vasc Biol. 2001; 21: 13-27.
Rea TJ, Brown CM, Sing CF. Complex adaptive system models and the genetic analysis of plasma HDL-cholesterol levels. Perspectives Biol Med. 2006;in press.
Hirschhorn JN, Daly MJ. Genome-wide association studies for common diseases and complex traits. Nature Rev Genet. 2005; 6: 95-108.
Reilly SL, Ferrell RE, Kottke BA, Kamboh MI, Sing CF. The gender-specific apolipoprotein E genotype influence on the distribution of lipids and apolipoproteins in the population of Rochester, MN I Pleiotropic effects on means and variances. Am J Hum Genet. 1991; 49: 1155-1166.
Lussier-Cacan S, Bolduc A, Xhignesse M, Nioysenga T, Connely PW, Sing CF. Impact of age and body size on inter-individual variation in measures of lipid metabolism: Influence of gender and apolipoprotein E genotype. Clin Genet. 2000; 57: 35-47.
Hughes GH, Cutter G, Donahue R, Friedman GD, Hulley S, Hunkeler E, Jacobs DR Jr., Liu K, Orden S, Pirie P, Tucker B, Wagenknecht LE. Recruitment in the Coronary Artery Disease Risk Development in Young Adults (Cardia) Study. Control Clin Trials. 1987; 8 (4 Suppl): 68S-73S.
Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR Jr., Liu K, Savage PJ. CARDIA: study design, recruitment, and some characteristics of the examined subjects. J Clin Epidemiol. 1988; 41: 1105-1116.
Warnick GR. Enzymatic methods for quantification of lipoprotein lipids. Methods Enzymol. 1986; 129: 101-123.
Warnick GR, Benderson J, Albers JJ. Dextran sulphate -Mg+ precipitation procedure for quantification of high density lipoprotein cholesterol. Clin Chem. 1982; 28: 1379-1388.
Friedewald WT, Levy RI, Frederickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma. Clin Chem. 1972; 18: 499-501.
Albers JJ, Adolphson JL. Comparison of commercial kits for apolipoprotein A-I and apolipoprotein B with standardized apolipoprotein A-I and B radioimmunoassay performed at the Northwest Lipid Research Center. J Lipid Res. 1988; 29: 102-108.
Sokal RR, Rohlf FJ. Biometry: the principles and practice of statistics in biological research. New York, NY: Freeman and Company; 1998.
Levene H. Robust tests for equality of variance. Contributions to probability and statistics. Stanford, CA: Stanford University Press; 1960.
Welch BL. On the comparison of several mean values: an alternative approach. Biometrika. 1951; 38: 330-336.
Excoffier L, Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol. 1995; 12: 921-927.
Fallin D, Schork NJ. Accuracy of haplotype frequency estimation for bialleleic loci, via the expectation-maximization algorithm for un-phased diploid genotype data. Am J Hum Genet. 2000; 67: 947-959.
Zaykin DV, Westfall PH, Young SS, Karnoub MA, Wagner MJ, Ehm MG. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered. 2001; 53: 79-91.
Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Royal Stat Soc. 1995; B57: 289-300.
Schmitz G, Kaminski WE, Orso E. ABC transporters in cellular lipid trafficking. Curr Opin Lipidol. 2000; 11: 493-501.
Schmitz G, Langmann T. Structure, function and regulation of the ABC1 gene product. Curr Opin Lipidol. 2001; 12: 129-140.
Frikke-Schmidt R, Nordestgaard BG, Jensen GB, Tybjærg-Hansen A. Genetic variation in ABC transporter A1 contributes to HDL cholesterol in the general population. J Clin Invest. 2004; 114: 1343-1353.
Clee SM, Zwinderman AH, Enghert JC, Zwarts KY, Molhuizen HOF, Roomp K, Jukema W, van Wijland M, vanDam M, Hudson TJ, Brooks-Wilson A, Genest J Jr, Kastelein JJP, Hayden M. Common genetic variation in ABCA1 is associated with altered lipoprotein levels and a modified risk for coronary artery disease. Circulation. 2001; 103: 1198-1205.
Lai C-Q, Parnell LD, Ordovas JM. The APOA1/C3/A4/A5 gene cluster, lipid metabolism and cardiovascular disease risk. Curr Opin Lipidol. 2005; 16: 153-166.
Klos KL, Hamon S, Clark AG, Boerwinkle E, Liu K, Sing CF. APOA5 polymorphisms influence plasma triglycerides in young, healthy African Americans and whites of the CARDIA Study. J Lipid Res. 2005; 46: 564-571.
Gibbons GH, Liew CC, Goodarzi MO, Rotter JI, Hsueh WA, Siragy HM, Pratt R, Dzau VJ. Genetic markers. Progress and potential for cardiovascular disease. Circulation. 2004; 109: IV47-IV58.
Jeenah M, Kessling A, Miller N, Humphries S. G to A substitution in the promoter region of the apolipoprotein AI gene is associated with elevated serum apolipoprotein AI and high density lipoprotein cholesterol concentrations. Mol Biol Med. 1990; 7: 233-241.
Kamboh MI, Bunker CH, Aston CE, Nestlerode CS, McAllister AE, Ukoli FA. Genetic association of five apolipoprotein polymorphisms with serum lipoprotein-lipid levels in African blacks. Genet Epidemiol. 1999; 16: 205-222. <a href="/cgi/external_ref?access_num=10.1002/(SICI)1098-2272(1999)16:2
Russo GT, Meigs JB, Cupples LA, Demissie S, Otvos JD, Wilson PWF, Lahoz C, Cucinotta D, Couture P, Mallory T, Schaefer EJ, Ordovas JM. Association of the Sst-I polymorphism at the APOC3 gene locus with variations in lipid levels, lipoprotein subclass profiles and coronary heart disease risk: the Framingham offspring study. Atherosclerosis. 2001; 158: 173-181.
Li G-P, Wang J-Y, Yan B-S, Xue H, Wu G. Genetic effect of two polymorphisms in the apolipoprotein A5 gene and apolipoprotein C3 gene on serum lipids and lipoproteins levels in a Chinese population. Clin Genet. 2004; 65: 470-476.
Hagberg JM, Wilund KR, Ferrell RE. APOE gene and gene-environment effects on plasma lipoprotein-lipid levels. Physiol Genomics. 2000; 4: 101-108.
Eichner JE, Dunn ST, Perveen G, Thompson DM, Stewart KE, Stroehla BC. apolipoprotein E polymorphism and cardiovascular disease: a HuGE review. Am J Epidemiol. 2002; 6: 487-495.
Ribalta J, Vallvé J-C, Girona J, Masana L. apolipoprotein and apolipoprotein receptor genes, blood lipids and disease. Curr Opin Clin Nutr Metab Care. 2003; 6: 177-187.
Stengård JH, Clark AG, Weiss KM, Kardia S, Nickerson DA, Salomaa V, Ehnholm C, Boerwinkle E, Sing CF. Contributions of 18 additional DNA sequence variations in the gene for apolipoprotein E to explaining variation in quantitative measures of lipid metabolism. Am J Hum Genet. 2002; 71: 501-517.
Boekholdt SM, Thompson JF. Natural variation as a tool in understanding the role of CETP in lipid levels and disease. J Lip Res. 2003; 44: 1080-1093.
A polymorphism and a novel promoter variant are independently associated with CETP concentration. Hum Mol Genet. 2003; 12: 111-123.
Thompson JF, Lira ME, Durham LK, Clark RW, Bamberger MJ, Milos PM. Polymorphisms in the CETP gene and association with CETP mass and HDL levels. Atherosclerosis. 2003; 167: 195-204.
Biegneux A, Hofmann AF, Young SG. Human CYP7A1 deficiency: progress and enigmas. J Clin Invest. 2002; 110: 109-117.
Wang J, Freeman DJ, Grundy SM, Levine DM, Guerra R, Cohen JC. Linkage between cholesterol 7a-hydroxylase and high plasma low-density lipoprotein cholesterol concentrations. J Clin Invest. 1998; 101: 1283-1291.
Lin JP, Myers RH, Almasy L, Coon HH, Arnett DK, Hong Y, Hunt SC. Linkage of the cholesterol 7-alpha-hydroxylase gene and low-density lipoprotein cholesterol conditional on apolipoprotein E association: the National Heart, Lung, and Blood Institute Family Heart Study Chin Med J. 2005; 118: 362-369.
Hofman MK, Groenendijk M, Verkuijlen PJJH, Jonkers IJAM, Mohrschladt MF, Smelt AHM, Princen HMG. Modulating effect of the A-278C promoter polymorphism in the cholesterol 7alph-hydroxylase gene on serum lipid levels in normolipidaemic and hypertriglyceridaemic individuals. Eur J Hum Genet. 2004; 12: 935-941.
Zhou B, Zhang SZ, Xiao CY, Zhang L, Li GX, Liu HK, Zheng KQ. [Association of 7alpha-hydroxylase gene polymorphism with levels of plasma lipids]. Yi Chuan. 2004; 26: 283-286.
Han Z, Heath SC, Shmulewitz D, Li W, Auerbach SB, Blundell ML, Lehner T, Ott J, Stoffel M, Friedman JM, Breslow JL. Candidate genes involved in cardiovascular risk factors by a family-based association study on the island of Kosrae, Federated States of Micronesia. Am J Med Genet. 2002; 110: 234-242.
Stein Y, Stein O. Lipoprotein lipase and atherosclerosis. Atherosclerosis. 2003; 170: 1-9.
Merkel M, Eckel RH, Goldberg IJ. Lipoprotein lipase: genetics, lipid uptake, and regulation. J Lipid Res. 2002; 43: 1997-2006.
Talmud PJ, Stephens JW. Lipoprotein lipase gene variants and the effect of environmental factors on cardiovascular disease risk. Diabetes Obesity Metab. 2004; 6: 1-7.
Nickerson DA, Taylor SL, Weiss KM, Clark AG, Hutchinson RG, Stengård J, Salomaa V, Vartiainen E, Boerwinkle E, Sing CF. SNA sequence diversity in a 9.7-kb region of the human lipoprotein lipase gene. Nature Genet. 1998; 19: 233-240.
Templeton AR, Clark AG, Weiss KM, Nickerson DA, Boerwinkle E, Sing CF. Recombinational and mutational hotspots within the human lipoprotein lipase gene. Am J Hum Genet. 2000; 66: 69-83.
Lee C-H, Olson P, Evans RM. Minireview: Lipid metabolism, metabolic diseases, and peroxisome proliferator-activated receptors. Endocrinology. 2003; 144: 2201-2207.
Chan E, Tan CS, Deurenberg-Yap M, Chia KS, Chew SK, Tai ES. The C227A polymorphism at the PPARA locus is associated with serum lipid concentrations and modulates the association between dietary polyunsaturated fatty acid intake and serum high density lipoprotein concentrations in Chinese women. Atherosclerosis. 2005; November 7. Epub Ahead of Print.
Liscum L, Underwood KW. Intracellular cholesterol transport and compartmentalization. J Biol Chem. 1995; 270: 15443-15446.
Hegele RA. Monogenic dyslipidemias: window on determinants of plasma lipoprotein metabolism. Am J Hum Genet. 69: 1161-1177.
Eberlé D, Hegarty B, Bossard P, Ferré P, Foufelle F. SREBP transcription factors: master regulators of lipid homeostasis. Biochimie. 2004; 86: 839-848.
Miserez AR, Muller PY, Barella L, Barella S, Staehelin HB, Leitersdorf E, Kark JD, Friedlander Y. Sterol-regulatory element-binding protein (SREBP)-2 contributes to polygenic hypercholesterolaemia. Atherosclerosis. 2002; 164: 15-26.
Duan X, Zhu W, Yin X, Dao J. [Relations between polymorphism of sterol regulatory element binding protein-2 gene and hyperlipidemia in the Han ethics of China]. Wei Sheng Yan Jiu. 2004; 33: 314-316.
作者单位：From Human Genetics Center (K.L.E.K., E.B., J.E.H.), University of Texas Health Science Center, Houston, Tex; Department of Human Genetics (C.F.S., T.J.R.), University of Michigan, Ann Arbor, Mich; Institute of Molecular Medicine (E.B., M.F.), University of Texas Health Science Center, Houston, Tex;