Sib-pair linkage analysis of longitudinal changes in lipoprotein risk factors and lipase genes in women twins
2000; Elsevier BV; Volume: 41; Issue: 8 Linguagem: Inglês
10.1016/s0022-2275(20)33438-6
ISSN1539-7262
AutoresYechiel Friedlander, Philippa J. Talmud, Karen L. Edwards, Steve E. Humphries, Melissa A. Austin,
Tópico(s)Lipid metabolism and biosynthesis
ResumoBased on longitudinal twin data in women, we have previously demonstrated a genetic influence on changes in lipoprotein risk factors, blood pressure measurements, and body mass index over a decade. The present study examined the linkage between changes in lipoprotein variables and candidate genes encoding the hormone-sensitive lipase (HSL), hepatic lipase (HL), and lipoprotein lipase (LPL). The sample consisted of 126 dizygotic (DZ) pairs of women twins who participated in the two examinations of the Kaiser Permanente Women Twins Study, performed a decade apart. Using quantitative sib-pair linkage analysis, a linkage was demonstrated between the locus for hormone-sensitive lipase and age-adjusted changes in plasma triglyceride (P = 0.015), which became more significant after adjustment for environmental factors and the exam-1 level (P = 0.005). There was also evidence suggesting linkage between the locus for hepatic lipase and changes in triglyceride (P = 0.023), but no linkage was detected for lipoprotein lipase and changes of lipid levels with time. These findings suggest that variation at these candidate gene loci may underlie a portion of the intraindividual variations in these coronary heart disease (CHD) risk factors, and that studies to identify the functional variants could provide new insights into genetic susceptibility to cardiovascular disease. —Friedlander, Y., P. J. Talmud, K. L. Edwards, S. E. Humphries, and M. A. Austin. Sib-pair linkage analysis of longitudinal changes in lipoprotein risk factors and lipase genes in women twins. J. Lipid Res. 2000. 41: 1302–1309. Based on longitudinal twin data in women, we have previously demonstrated a genetic influence on changes in lipoprotein risk factors, blood pressure measurements, and body mass index over a decade. The present study examined the linkage between changes in lipoprotein variables and candidate genes encoding the hormone-sensitive lipase (HSL), hepatic lipase (HL), and lipoprotein lipase (LPL). The sample consisted of 126 dizygotic (DZ) pairs of women twins who participated in the two examinations of the Kaiser Permanente Women Twins Study, performed a decade apart. Using quantitative sib-pair linkage analysis, a linkage was demonstrated between the locus for hormone-sensitive lipase and age-adjusted changes in plasma triglyceride (P = 0.015), which became more significant after adjustment for environmental factors and the exam-1 level (P = 0.005). There was also evidence suggesting linkage between the locus for hepatic lipase and changes in triglyceride (P = 0.023), but no linkage was detected for lipoprotein lipase and changes of lipid levels with time. These findings suggest that variation at these candidate gene loci may underlie a portion of the intraindividual variations in these coronary heart disease (CHD) risk factors, and that studies to identify the functional variants could provide new insights into genetic susceptibility to cardiovascular disease. —Friedlander, Y., P. J. Talmud, K. L. Edwards, S. E. Humphries, and M. A. Austin. Sib-pair linkage analysis of longitudinal changes in lipoprotein risk factors and lipase genes in women twins. J. Lipid Res. 2000. 41: 1302–1309. A number of studies have demonstrated significant associations between changes in coronary heart disease (CHD) risk factors and subsequent risk of CHD (1Groover M.E. Jernigan J.A. Martin C.D. Variations in serum lipid concentration and clinical coronary disease.Am. J. Med. Sci. 1960; 53: 133-139Google Scholar, 2Hamm P.B. Schekelle R.B. Stamler J. Large fluctuation in body weight during young adulthood: twenty-five-year risk of coronary death in men.Am. J. Epidemiol. 1989; 129: 312-318Google Scholar, 3Lissner L. Odell P.M. D'Agostino III, R.B. Stokes J. Kreger B.E. Belanger A.J. Brownell K.D. Variability of body weight and health outcomes in the Framingham population.N. Engl. J. Med. 1991; 324: 1839-1844Google Scholar). It has been proposed that genetic–environmental interactions could reflect the presence of "variability genes." Such variability genes are influenced by environmental factors and may, in turn, cause risk factor variation over time in response to the environmental change (4Berg K. Gene–environment interaction: variability gene concept.in: Goldbourt U. de Faire U. Berg K. Genetic Factors in Coronary Heart Disease. Kluwer Academic Publishers, Dordrecht1994: 373-383Google Scholar, 5Fabsitz R.R. Sholinsky P. Carmelli D. Genetic influences on adult weight gain and maximum body mass index in male twins.Am. J. Epidemiol. 1994; 140: 711-720Google Scholar, 6Austin M.A. Friedlander Y. Newman B. Edwards K. Mayer E.J. King M-C. Genetic influences on changes in body mass index: a longitudinal analysis of women twins.Obes. Res. 1997; 5: 326-331Google Scholar, 7Friedlander Y. Austin M.A. Newman B. Edwards K. Mayer-Davis E.J. King M-C. Heritability of longitudinal changes in coronary heart disease risk factors in women twins.Am. J. Hum. Genet. 1997; 60: 1502-1512Google Scholar). The variability gene effect can be understood as the idea that some genes are "switched on" in response to specific environmental factors, remain continuously active, and thus regulates the individual's phenotypic response to the ongoing environmental exposure. "Level genes," on the other hand, exhibit an association with absolute risk factor levels. In a longitudinal observation, changes over time probably reflect elements of both processes. Lipoprotein lipase (LPL) is a major enzyme responsible for the hydrolysis of triglycerides in chylomicrons and very low density lipoproteins (VLDL) (8Bensadoun A. Lipoprotein lipase.Annu. Rev. Nutr. 1991; 11: 217-237Google Scholar). It also promotes the exchange of lipids between VLDL and high density lipoproteins (HDL) (9Eckel R.H. Lipoprotein lipase. A multifunctional enzyme relevant to common metabolic diseases.N. Engl. J. Med. 1989; 320: 1060-1068Google Scholar) and is an essential enzyme in the formation of low density lipoprotein (LDL) particles (10Zambon A. Hokanson J.E. Gagne C. Lupien P.J. Moorjani S. Hayden M.R. Brunzell J.D. Combined effect of familial hypercholesterolemia and lipoprotein lipase deficiency in determining low-density-lipoprotein size buoyancy, and lipid composition.J. Invest. Med. 1996; 44: A91Google Scholar). A meta-analysis that combined data of 14 studies, representing 15,000 subjects, has shown that allelic variants in the LPL gene were associated with coronary heart disease risk (11Hokanson J.E. Lipoprotein lipase gene variants and risk of coronary disease: a quantitative analysis of population-based studies.Int. J. Clin. Lab. Res. 1997; 27: 24-34Google Scholar). More than 70 rare mutations in the LPL gene have been reported in families with inherited LPL deficiency (12Murthy V. Julien P. Gagne C. Molecular pathobiology of the human lipoprotein lipase.Pharmacol. Ther. 1996; 70: 101-135Google Scholar). In addition to these, several common variants and genetic polymorphisms have been identified with a more modest effect on LPL catalytic function. The first common mutation described in a coding sequence was Ser447X (exon 9), caused by a C–G transversion that results in the substitution of Ser-447 by a premature stop codon and leads to the truncation of LPL (13Hata A. Robertson M. Emi M. Lalouel J.M. Direct detection and automated sequencing of individual alleles after electrophoretic strand separation: identification of a common nonsense mutation in exon 9 of the human lipoprotein lipase gene.Nucleic Acids Res. 1990; 18: 5407-5411Google Scholar). This variant, which does not seem to alter LPL lipolytic activity yet increases the secretion of LPL mass (14Zhang H. Henderson H. Gagne S.E. Clee S.M. Miao L. Liu G. Hayden M.R. Common sequence variants of lipoprotein lipase: standardized studies of in vitro expression and catalytic function.Biochim. Biophys. Acta. 1996; 1302: 159-166Google Scholar), was found to be associated with lower triglyceride levels (15Stocks J. Thorn J.A. Galton D.J. Lipoprotein lipase genotypes for a common premature termination colon mutation detected by PCR-mediated site-directed mutagenesis and restriction digestion.J. Lipid Res. 1992; 33: 853-857Google Scholar) and with a lower risk of CHD (16Jemaa R. Tuzet S. Portos C. Betoulle D. Apfelbaum M. Fumeron F. Lipoprotein lipase gene polymorphism: associations with hypertriglyceridemia and body mass index in obese people.Int. J. Obes. 1995; 19: 270-274Google Scholar). In one study 18.4% of the patients with coronary artery disease carried at least one copy of the Ser447X allele (17Groenemeijer B.E. Hallman M.D. Reymer P.W.A. Gagne E. Kuivenhoven J.A. Bruin T. Jansen H. Lie K.I. Bruschke A.V.G. Boerwinkle E. Hayden M.R. Kastelein J.J.P. on behalf of the REGRESS Study Group Genetic variant showing a positive interaction with β-blocking agents with beneficial influence on lipoprotein lipase activity, HDL cholesterol, and triglyceride levels in coronary artery disease patients. The Ser447Stop substitution in the lipoprotein lipase.Circulation. 1997; 95: 2628-2635Google Scholar). Carriers of this common variant showed significantly higher LPL activity and HDL cholesterol (HDL-C) levels and lower triglyceride levels than noncarriers. Other investigators, however, did not identify associations of the Ser447X with differences in plasma lipid levels, concluding that it is a neutral polymorphism (18Peacock R.E. Hamsten A. Nilsson-Ehle P. Humphries S.E. Associations between lipoprotein lipase gene polymorphisms and plasma correlations of lipids, lipoproteins and lipase activities in young myocardial infarction survivors and age-matched health individuals from Sweden.Atherosclerosis. 1992; 97: 171-185Google Scholar). Although no convincing epidemiological data are available on the association between hepatic lipase (HL) and CHD it plays a major role in the control of several lipoproteins (19Kinnunen P.H.J. Virtanen J.A. Vainio P. Lipoprotein lipase and hepatic endothelial lipase: their roles in plasma lipoprotein metabolism.Atheroscler. Rev. 1983; 11: 65-105Google Scholar, 20Applebaum-Bowden D. Lipases and lecithin: cholesterol acyltransferase in the control of lipoprotein metabolism [review].Atheroscler. Rev. 1995; 11: 65-105Google Scholar). HL is responsible for the lipolysis of VLDL remnant particles as well as the conversion of larger HDL subclass 2 (HDL2) to smaller HDL3 particles. Studies have shown that HL is also involved in lipolysis of large, buoyant LDL and high HL activity is associated with an increase in small, dense LDL particles (21Zambon A. Austin M.A. Brown B.G. Hokanson J.E. Brunzell J.D. Effect of hepatic lipase on LDL in normal men and those with coronary artery disease.Arterioscler. Thromb. 1993; 13: 147-153Google Scholar, 22Watson T.D.G. Caslake M.J. Freeman D.J. Griffin B.A. Hinnie J. Packard C.J. Shepherd J. Determinants of LDL subfraction distribution and concentrations in young normolipidemic subjects.Arterioscler. Thromb. 1994; 14: 902-910Google Scholar). Genetic polymorphisms in the promoter region of the HL gene were found to be strongly associated with the observed variation in HL activity (23Zambon A. Deeb S.S. Hokanson J.E. Brown B.G. Brunzell J.D. Common variants in the promoter region of the hepatic lipase gene are associated with lower levels of hepatic lipase activity, buoyant LDL, and higher HDL2 cholesterol.Arterioscler. Thromb. Vasc. Biol. 1998; 18: 1723-1729Google Scholar) and HDL-C (24Cohen J.C. Wang Z. Grundy S.M. Stoesz M.R. Guerra R. Variation at the hepatic lipase and apolipoprotein AI/CIII/AIV loci is a major cause of genetically determined variation in plasma HDL cholesterol levels.J. Clin. Invest. 1994; 94: 2377-2384Google Scholar, 25Guerra R. Wang J. Grundy S.M. Cohen J.C. A hepatic lipase (LIPC) allele associated with high plasma concentrations of high density lipoprotein cholesterol.Proc. Natl. Acad. Sci. USA. 1997; 94: 4532-4537Google Scholar). In a case-control study of young men with parental history of premature myocardial infarction and age-matched controls, the HL promoter variant C-480T was found to be associated also with concentration of plasma triglyceride and total cholesterol (26Jansen H. Chu G. Ehnholm C. Dallongeville J. Nicaud V. Talmud P.J. for the EARS Group The T allele of the hepatic lipase promoter variant C–480T is associated with increased fasting lipids and HDL and increased preprandial and postprandial LpCIII:B. European Atherosclerosis Research Study (EARS) II.Arterioscler. Thromb. Vasc. Biol. 1999; 19: 303-308Google Scholar). A preliminary study showed that the missense mutation Val73Met of the HL gene was overrepresented in 25 German subjects with familial combined hyperlipidemia (FCHL) compared with controls (27Gehrisch S. Tesche R. Kostka H. Julius U. Jaross W. Point mutations in the hepatic triglyceride lipase (HTGL) gene in familial combined hyperlipidemia (FCHL).Circulation. 1995; 92: 493Google Scholar). Yet, in 14 well-documented Finnish pedigrees with premature CHD and FCHL there was no evidence of linkage between FCHL and the HL gene (28Pajukanta P. Porkka K.V. Antikainen M. Taskinen M.R. Perola M. Murtomaki-Repo S. Ehnholm S. Nuotio I. Suurinkeroinen L. Lahdenkari A.T. Syvanen A.C. Viikari J.S. Ehnholm C. Peltonen L. No evidence of linkage between familial combined hyperlipidemia and genes encoding lipolytic enzymes in Finnish families.Arterioscler. Thromb. Vasc. Biol. 1997; 17: 841-850Google Scholar). Hormone-sensitive lipase (HSL) has a key role in the regulation of lipolysis in adipocytes by catalyzing the breakdown of triglyceride to glycerol and free fatty acids (29Stralfors P. Olsson H. Belfrage P. Hormone-sensitive lipase.in: Boyer P.D. Krebs E.G. The Enzymes. 18. Academic Press, New York1987: 147-177Google Scholar, 30Yeaman S.J. Hormone-sensitive lipase—a multipurpose enzyme in lipid metabolism.Biochim. Biophys. Acta. 1990; 1052: 128-132Google Scholar). In addition, HSL has cholesterol hydrolase activity and generates free cholesterol for steroid synthesis in steroidogenic tissue (31Holm C. Kirchgessner T.G. Svenson K.L. Fredrikson G. Nilsson S. Miller C.G. Shively J.E. Heinzmann C. Sparkes R.S. Mohandas T. Hormone-sensitive lipase: sequence, expression, and chromosomal localization to 19 cent-q13.3.Science. 1988; 16: 1503-1506Google Scholar) and macrophages (32Reue K. Cohen R.D. Schotz M.C. Evidence for hormone-sensitive lipase mRNA expression in human monocyte/macrophages.Arterioscler. Thromb. Vasc. Biol. 1997; 17: 3428-3432Google Scholar). In a small study of type II diabetics and health controls, variation in the HSL gene was shown to be associated with total cholesterol but not with triglyceride (33Shimada F. Makino H. Hashimoto N. Iwaoka H. Taira M. Nozaki O. Kanatsuka A. Holm C. Langin D. Saito Y. Detection of an amino acid polymorphism in hormone-sensitive lipase in Japanese subjects.Metabolism. 1996; 45: 862-864Google Scholar). Previous work has suggested that in FCHL there is a lipolytic defect due to impaired function of HSL (34Reynisdottir S. Eriksson M. Angelin B. Arner P. Impaired activation of adipocyte lipolysis in familial combined hyperlipidemia.J. Clin. Invest. 1995; 95: 2161-2169Google Scholar, 35Reynisdottir S. Angelin B. Langin D. Lithell H. Eriksson M. Holm C. Arner P. Adipose tissue lipoprotein lipase and hormone-sensitive lipase. Contrasting findings in familial combined hyperlipidemia and insulin resistance syndrome.Arterioscler. Thromb. Vasc. Biol. 1997; 17: 2287-2292Google Scholar). However, no linkage was found between the well-documented FCHL Finnish pedigrees and the HSL gene (28Pajukanta P. Porkka K.V. Antikainen M. Taskinen M.R. Perola M. Murtomaki-Repo S. Ehnholm S. Nuotio I. Suurinkeroinen L. Lahdenkari A.T. Syvanen A.C. Viikari J.S. Ehnholm C. Peltonen L. No evidence of linkage between familial combined hyperlipidemia and genes encoding lipolytic enzymes in Finnish families.Arterioscler. Thromb. Vasc. Biol. 1997; 17: 841-850Google Scholar). The purpose of this study was to determine whether there was any evidence of genetic linkage between longitudinal changes in lipoprotein risk factors levels and the loci at genes encoding three lipases known to be involved in lipid metabolism. The study makes use of a large sample of dizygotic adult women twins who participated in two examinations of the Kaiser Permanente Women Twins Study a decade apart, using quantitative sib-pair linkage analysis. The women twins included in this analysis participated in both the first and second examinations of the Kaiser Permanente Women Twins Study in Oakland, California, conducted during 1978–1979 and 1989–1990, respectively. The sample selection, protocol for data collection, laboratory methodology, and other basic features of this study have been described in detail elsewhere (36Austin M.A. King M-C. Bawol R.D. Hulley S.B. Friedman G.D. Risk factors for coronary heart disease in adult female twins. Genetic heritability and shared environmental influences.Am. J. Epidemiol. 1987; 125: 308-317Google Scholar, 37Selby J.V. Austin M.A. Newman B. Zhang D. Quesenberry C.P. Mayer E.J. Krauss R.M. LDL subclass phenotypes and the insulin resistance syndrome.Circulation. 1993; 88: 381-387Google Scholar). Briefly, the study included 203 monozygotic (MZ) and 145 dizygotic (DZ) twin pairs with mean ages of 41 and 51 years at examinations 1 and 2, respectively. At both examinations, lipid and lipoprotein determinations were based on blood samples drawn after a 12-h fast. LDL cholesterol was estimated according to the Friedwald formula (38Friedwald W.T. Levy R.I. Fredrickson D.S. Estimation of the concentration of low density lipoprotein cholesterol in plasma—without use of preparative ultracentrifuge.Clin. Chem. 1972; 18: 449-502Google Scholar). Subjects and their co-twins were excluded from the lipid analyses, if at either examination, either one of each pair was not fasting, was taking lipid-altering medications, or had missing or extreme values (total cholesterol [TC] >350 mg/dL or triglyceride [TG] >400 mg/dL). Height and weight were measured at both examinations while subjects were dressed in lightweight clothes with shoes removed. The study was approved by the Kaiser Permanente Institutional Review Board and each woman provided written informed consent for participation in the study. Genomic DNA was prepared from whole blood after lysis of red blood cells and of the 290 available blood samples from DZ twins, DNA extraction succeeded for 280 (96.6%). The three candidate genes examined in this analysis, and their chromosomal locations, are listed in Table 1 (39Austin M.A. Talmud P.J. Luong L-A. Haddad L. Day I.N.M. Newman B. Edwards K.L. Krauss R.M. Humphries S.E. Candidate-gene studies of the atherogenic lipoprotein phenotype: a sib-pair linkage analysis of DZ women twins.Am. J. Hum. Genet. 1998; 62: 406-419Google Scholar, 40Humphries S.E. Nicaud V. Margalef J. Tiret L. Talmud P.J. for the EARS Group Lipoprotein lipase gene variation is associated with a paternal history of premature coronary artery disease and fasting and postprandial plasma triglycerides.Arterioscler. Thromb. Vasc. Biol. 1998; 18: 526-534Google Scholar, 41Bhattacharya S. Ameis D. Cullen P. Norcisi T.M. Bayliss J. Greten H. Schotz M.C. Scott J. VNTR polymorphism in the hepatic lipase gene (LIPC).Nucleic Acids Res. 1991; 19: 5088Google Scholar, 42Levitt R.C. Jedlicka A.E. Nouri N. Dinucleotide repeat polymorphism at the hormone sensitive lipase (LIPE) locus.Hum. Mol. Genet. 1992; 1: 139Google Scholar). These genes included the hepatic lipase (HL) and hormone-sensitive lipase (HSL). In addition, we identify carriers with the Ser447X variant at the LPL gene. Polymerase chain reaction (PCR) analysis was performed using primers 5′-CATCCATTTTCTTCCACAGGG-3′ (sense) and 5′-GCCCAGAATGCTCACCAGACT-3′ (antisense). After amplification, the PCR product (137 bp) was digested with HinfI and the fragments separated by 7.5% microtiter array diagonal gel electrophoresis (MADGE) (43Day I.N.M. Humphries S.E. Electrophoresis for genotyping: microtitre array diagonal gel electrophoresis (MADGE) on horizontal polyacrylamide (H-PAGE) gels, Hydrolink or agarose.Anal. Biochem. 1994; 222: 389-395Google Scholar). The Stop447 allele after Hinf I digestion gives two fragments, 114 and 23 bp in length.TABLE 1.Candidate gene polymorphisms typed for women twinsaTable reproduced in part from Austin et al. (39).Candidate GeneChromosomeMarkerNo. of AllelesHeterozygosity IndexbBased individually on all women in the study.ReferenceLPL8HinfI20.17Humphries et al. (40Humphries S.E. Nicaud V. Margalef J. Tiret L. Talmud P.J. for the EARS Group Lipoprotein lipase gene variation is associated with a paternal history of premature coronary artery disease and fasting and postprandial plasma triglycerides.Arterioscler. Thromb. Vasc. Biol. 1998; 18: 526-534Google Scholar)HL15CA repeat in intron 870.63Bhattacharya et al. (41Bhattacharya S. Ameis D. Cullen P. Norcisi T.M. Bayliss J. Greten H. Schotz M.C. Scott J. VNTR polymorphism in the hepatic lipase gene (LIPC).Nucleic Acids Res. 1991; 19: 5088Google Scholar)HSL19GT repeat in intron 7140.67Levitt et al. (42Levitt R.C. Jedlicka A.E. Nouri N. Dinucleotide repeat polymorphism at the hormone sensitive lipase (LIPE) locus.Hum. Mol. Genet. 1992; 1: 139Google Scholar)a Table reproduced in part from Austin et al. (39Austin M.A. Talmud P.J. Luong L-A. Haddad L. Day I.N.M. Newman B. Edwards K.L. Krauss R.M. Humphries S.E. Candidate-gene studies of the atherogenic lipoprotein phenotype: a sib-pair linkage analysis of DZ women twins.Am. J. Hum. Genet. 1998; 62: 406-419Google Scholar).b Based individually on all women in the study. Open table in a new tab Genotypes for the three genes were determined for a total of 126 DZ twin pairs. However, because both co-twins are needed for this analysis, pairs were excluded if a genotype could not obtained for one or both co-twins in a pair. Thus, the sample sizes varied for specific candidate genes and the phenotype examined. Prior to the linkage analyses, the changes in the risk factors between examinations were first adjusted for age, using regressionanalysis. Because co-twins in the same pair tend to share environment as well as genes, in addition to the age adjustment, changes in risk factor levels were adjusted for available environmental and behavioral variables at each of the two examinations (7Friedlander Y. Austin M.A. Newman B. Edwards K. Mayer-Davis E.J. King M-C. Heritability of longitudinal changes in coronary heart disease risk factors in women twins.Am. J. Hum. Genet. 1997; 60: 1502-1512Google Scholar). Finally, the risk factor value at examination 1 was also included in the regression models to adjust for potential effects of regression to the mean. For details of the adjustment procedure see Friedlander et al. (7Friedlander Y. Austin M.A. Newman B. Edwards K. Mayer-Davis E.J. King M-C. Heritability of longitudinal changes in coronary heart disease risk factors in women twins.Am. J. Hum. Genet. 1997; 60: 1502-1512Google Scholar) Linkage between the polymorphisms at various gene loci and the observed variation in changes of risk factors were estimated by using a sib-pair linkage procedure implemented in the SAGE (Statistical Analysis for Genetic Epidemiology) package (44SAGE Statistical Analysis for Genetic Epidemiology. Release 2.2. Department of Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, OH1994Google Scholar). The underlying basis for this approach is to compare the quantitative variation in the changes of risk factors between co-twins as a function of the number of marker alleles that the twin pair shares identically by descent (IBD) (45Haseman J.K. Elston R.C. The investigation of linkage between a quantitative trait and a marker locus.Behav. Genet. 1972; 2: 2-19Google Scholar). Because parental data are not available in this twin study, IBD was estimated from the observed identity by state (IBS) and based on the frequencies of alleles for each marker (46Amos, C. I., 1988. Robust Method for Detection of Genetic Linkage for Data from Extended Families and Pedigrees. PhD Dissertation. Louisiana State University Medical Center, New Orleans, LA.Google Scholar, 47Amos C.I. Dawson D.V. Elston R.C. The probabilistic determination of identity-by-descent sharing for pairs of relatives from pedigrees.Am. J. Hum. Genet. 1990; 47: 842-853Google Scholar). As in Austin et al. (39Austin M.A. Talmud P.J. Luong L-A. Haddad L. Day I.N.M. Newman B. Edwards K.L. Krauss R.M. Humphries S.E. Candidate-gene studies of the atherogenic lipoprotein phenotype: a sib-pair linkage analysis of DZ women twins.Am. J. Hum. Genet. 1998; 62: 406-419Google Scholar), allele frequencies were estimated from the same sample of DZ twins, because no population-based allele frequencies are currently available for several of the markers used in this study. Allele frequencies for the entire sample were used, because they were similar to frequencies based on one randomly selected co-twin per pair. Because we used a candidate gene approach and in order to reduce the possibility of type I error, an α value of 0.05 has been used as the criterion for statistical significance. The adjusted P value (Padjusted), correcting for multiple number of genes examined, is also reported for each result (48Finner H. Some new inequalities for the range distribution with application to the determination of optimum significance levels of multiple range tests.J. Am. Stat. Assoc. 1990; 85: 191-194Google Scholar). As a supplement to the sib-pair method, when evidence of linkage to a specific marker was found, intraclass correlation of the appropriate phenotype was calculated within twin pairs stratified into those sharing 2, 1, and 0 alleles IBS. When linkage is present, twins sharing 2 alleles IBS are expected to have higher correlation than those sharing 1, which in turn should be higher that those pairs sharing 0 alleles. Because the standard errors of intraclass correlations are difficult to compute, P values are reported on the basis of interclass correlations with the same sample size. Results of the sib-pair linkage analysis for change in lipid and lipoprotein variables, adjusted for age, environment,and baseline values, provided evidence of linkage between HSL and HL with ΔTG (Table 2). As shown in Fig. 1, the slope from the regression of the estimated proportion of alleles identical by descent at the HSL locus versus the co-twin differences in adjusted ΔTG squared has a significantly negative slope (b = −2.297; P = 0.005; Padjusted = 0.149), demonstrating evidence of linkage between HSL and ΔTG. In addition, the analysis showed evidence of linkage of the HL gene to adjusted ΔTG (b = −2.108; P = 0.023; Padjusted = 0.034) (Fig. 2). No evidence of linkage was obtained for adjusted changes in lipid variables and LPL.TABLE 2.Slope and P values from quantitative sib-pair linkage analysis on DZ women twins for age, environment, and baseline-adjusted changes in lipid and lipoprotein variables and hormone-sensitive lipase (HSL), hepatic lipase (HL), and lipoprotein lipase (LPL) genesLipid VariableNo. of PairsRegression Slope (P Value)HSLHLLPLTotal cholesterol1081.307 (0.989)3.322 (0.981)0.053 (0.512)LDL cholesterol108−0.177 (0.432)1.531 (0.884)2.121 (0.792)Triglyceride108−2.297 (0.005)−2.108 (0.023)−2.124 (0.251)HDL cholesterol108−1.077 (0.163)1.226 (0.823)2.948 (0.888) Open table in a new tab Fig. 2.Results of quantitative sib-pair linkage analysis of the HL gene in DZ women twins. The x axis indicates the estimated number of alleles shared by co-twins in a pair, at the HL marker. The y axis indicates the squared co-twin difference in age, environment, and baseline-adjusted ΔTG level (in mg/dl). Note the significant negative slope (b = −2.108; P = 0.023) of the regression line, which is consistent with evidence of genetic linkage.View Large Image Figure ViewerDownload (PPT) The separate and combined effects of variation in HSL and HL loci were also examined. The distributions of squared twin-pair differences of TG changes among twins sharing different numbers of HSL and HL alleles IBS are presented in Fig. 3. The disparity in the distribution of squared twin-pair differences of change in TG among twins sharing no HL alleles IBS was clearly greater than among those twins sharing 1 or 2 HL alleles IBS. Similarly, the distribution of squared twin-pair differences among twins sharing 0 or 1 HSL alleles IBS was considerably greater than among those sharing both HSL alleles IBS. Finally, the combined effects of variation at HL and HSL loci is presented when twin pairs were assigned to sharing 0–2, 3, and 4 alleles at both loci. A clear monotonic decrease in means and standard errors of squared twin-pair differences of ΔTG with an increase in the number of alleles of these genes shared by twins is indicated. An alternative way to illustrate the linkage of a gene controlling a quantitative trait and a genetic marker is to compare the correlations among those twin pairs sharing 2, 1, or 0 alleles IBS, thus avoiding estimation of IBD status. As expected from the linkage data, we obtained a strong positive correlation for ΔTG, r = 0.33 among those twin pairs sharing both HSL alleles IBS and in contrast, a negative correlation, r = −0.24, among those twin pairs sharing 0 HSL alleles IBS (Fig. 4). Similarly, the intraclass correlations for age-, environment-, and baseline level-adjusted ΔTG were highest among pairs sharing 2 alleles at the HL gene and considerably lower for those sharing 1 allele or 0 alleles (Fig. 5).Fig. 5.Similarity of co-twin values for adjusted ΔTG, among pairs sharing (A
Referência(s)