Common Genomic Variants Associated With Variation in Plasma Lipoproteins in Young Aboriginal Canadians
Abstract We hypothesized that common genomic variants would be associated with variation in lipoprotein phenotypes in young subjects. We determined genotypes of FABP2, PON, APOC3, and APOE in 188 aboriginal Canadians, aged 9 to 17 years. We found that 13 of 32 possible genotype-phenotype associations were significant: (1) the FABP2 codon 54 genotype was associated with variation in plasma triglycerides (P=.045); (2) the PON codon 192 genotype was associated with variation in plasma total and LDL cholesterol and apoB (P=.0099, P=.0088, and P=.016, respectively); (3) the APOC3 insulin-response-element genotype was associated with variation in plasma triglycerides, HDL cholesterol, apoA-I, the total cholesterol to HDL cholesterol ratio, and the apoB to apoA-I ratio (P=.0014, P=.0069, P=.045, P=.0021, and P=.0081, respectively); and (4) the APOE restriction isotype was associated with variation in plasma LDL cholesterol, apoB, the total cholesterol to HDL cholesterol ratio, and the apoB to apoA-I ratio (P=.025, P=.034, P=.045, and P=.047, respectively). The average young age and relative absence of age-dependent secondary environmental factors could have eased the identification of small genetic effects on lipoprotein phenotypes in this study sample.
- Received June 12, 1996.
- Accepted August 30, 1996.
Atherosclerosis begins in childhood and progresses throughout life.1 The childhood lesions of atherosclerosis are strongly associated with risk factors that predispose to CHD in adults.2 3 In particular, abnormal plasma lipids and lipoproteins in youths are associated with pathological end points, such as fatty streaks in the aorta and coronary arteries and fibrous plaques in the coronary arteries.4 Genetic factors are considered to be important determinants of plasma lipoproteins in adults.5 6 7 The role of genetic factors in determining lipoprotein phenotypes in children and adolescents is established for uncommon monogenic conditions such as familial hypercholesterolemia or familial lipoprotein lipase deficiency.2 However, the role of genetics in determining plasma lipoproteins in children and adolescents is not clear. In young subjects, the absence of exposure to many secondary factors, such as aging, obesity, poor lifestyle, and the influence of sex hormones, should permit a cleared definition of genetic determinants, which individually would be expected to have small effects on plasma lipoproteins.
There have been numerous reports of associations between DNA markers of candidate genes and lipoprotein metabolism.5 6 7 However, alleles of only a few genes appear to be consistently related to lipoprotein phenotypes.6 7 For example, the E4 allele of APOE gene is associated with higher mean plasma concentrations of LDL cholesterol, while the E2 allele is associated with lower mean plasma concentrations of LDL cholesterol, than the E3 allele in most populations.8 This genomic variation underlies structural variation in apoE, which is felt to underlie the association with the metabolic phenotypes.6 8 The T54 allele of the FABP2 gene has been associated with diabetes-related phenotypes in Pima Indians9 and Mexican-Americans.10 This genomic variation underlies variation of in vitro protein function, which is felt to underlie the association with the metabolic phenotypes.9 The R192 allele of the PON gene has been associated with a proatherogenic lipoprotein profile11 and with increased risk of CHD.12 13 This genomic variation underlies variation of paraoxonase activity14 but may not underlie the phenotype-genotype associations. Markers of the APOC3 gene have been associated with variation in plasma TG.15 16 17 One common APOC3 genomic variant contains a C at −455 nt within the insulin-response element of the APOC3 promoter. The C-containing variant was associated with the loss of insulin-mediated downregulation of APOC3 gene expression.18 This genomic variation was proposed to underlie the association between APOC3 gene polymorphism and lipoprotein variation.18
We wished to determine whether genomic variation in FABP2, PON, APOC3, and APOE would be associated with variation in plasma lipoproteins in young subjects. We examined the association between plasma lipoproteins and genomic variation in FABP2, PON, APOC3, and APOE in aboriginal Canadians aged 17 years and younger.
The community of Sandy Lake, Ontario is located 2000 km northwest of Toronto, in the subarctic boreal forest of central Canada. This isolated community is accessible only by air during most of the year. The ancestors of the contemporary residents lived a nomadic, hunting-gathering subsistence typical of Algonkian-speaking peoples of the northeastern subarctic. Since the development of the reservation and residential school systems, the lifestyle has changed from extremely physically active to extremely sedentary. The primary food source has changed from wildlife, roots, and berries to processed high-fat foods.
One hundred eighty-eight community members, aged 17 years and younger, participated in the study. Assessments included clinical evaluation. PBF was estimated by bioelectrical impedance analysis using the Tanita TBF-201 body fat analyzer (Tanita Corporation).19 20 Volunteers gave blood samples after fasting overnight for 8 to 12 hours. Exclusion criteria included an inadequate blood sample for all biochemical and/or genetic determinations. The project was approved by the University of Toronto Ethics Review Committee.
Biochemical and Genetic Analyses
Blood for plasma lipoprotein analyses was centrifuged at 2000 rpm for 30 minutes and the serum was stored at −70°C. Concentrations of lipids, lipoproteins, and apolipoproteins were determined as described.21 22 23 Established procedures were used to extract leukocyte DNA and to determine genotypes of FABP2 codon 54,9 PON codon 192,14 and APOE exon 4.24 A novel gene amplification reaction was devised to genotype APOC3. The primers C3-pro-5′ (5′-GTGAGAGCTCAGCCCCTGTAA-3′) and C3-pro-3′ (5′-TTTCACACTGGAAATTTCAGG-3′) were used in a gene-amplification program with annealing temperature 60°C to amplify a 194-bp fragment of the APOC3 promoter that contained the insulin-response element. Digestion of a fragment that derived from the wild-type, or insulin-responsive, promoter sequence with endonuclease Fok I resulted in two fragments with sizes 122 and 72 bp. Digestion of a fragment that derived from an allele with the substitution of C for T at position −455 nt with endonuclease Fok I resulted in a single fragment with size 194 bp. All genotyping reactions were run with known controls.
The significance of deviations of observed genotype frequencies from those predicted by the Hardy-Weinberg equation were evaluated with χ2 tests. SAS (version 6.08) was used for all statistical comparisons involving biochemical variables.25 For parametric statistical analyses, each quantitative variable was transformed and subjected to analysis of normality as described.21 22 23 All untransformed biochemical variables had distributions that were significantly nonnormal. Transformation using the ln for each variable resulted in either a distribution that was nonsignificantly different from normal or a distribution that was markedly more normal than that for the untransformed variable (data not shown).
ANOVA was performed, using the general linear models procedure to determine the sources of variation for fasting plasma TC, TG, LDL cholesterol, HDL cholesterol, apoB and apoA-I, and the TC/HDL and apoB/A-I ratios, with F tests computed from the type III sums of squares.25 This form of sums of squares is applicable to unbalanced study designs. Independent variables for each ANOVA were sex, age, and the ln of PBF. Also included as independent variables for ANOVA were four genotypes: FABP2 codon 54, PON codon 192, APOC3 position −455 nt, and APOE restriction isotype. When a significant association with a genotype was found, differences in least-squares means for the plasma lipoprotein trait were compared between genotypes.25
Sufficient DNA and phenotypic information were obtained for analysis from 188 subjects aged 17 years and below, of whom 122 (62.2%) were female. Mean values and ranges for clinical and biochemical traits are shown in Table 1⇓. There were no significant differences between males and females for any of the biochemical traits (data not shown).
Allele and Genotype Frequencies
The frequencies of the alleles of the four genotype systems used in the subsequent ANOVA are shown in Table 2⇓. Some allele frequencies differed markedly from those reported in other populations. Specifically, the allele frequency of the PON 192Q was 0.240, which was very low compared with the frequency of ≈0.65 reported in white study samples.11 12 13 14 Also, the frequencies of both the E4 and E2 alleles of APOE were much lower than in other populations studied, including Greenland Inuit.26 Observed genotype frequencies did not deviate from those predicted by the Hardy-Weinberg equation (all P>.10).
The results of the ANOVA are shown in Table 3⇓. Since ANOVA takes multiple comparisons into account, we did not have to adjust the levels of nominal significance. Sex was not significantly associated with variation in any biochemical trait. Age was significantly associated with variation in the logarithms of LDL and HDL cholesterol, apoB, TC/HDL ratio, and apoB/A-I ratio. The logarithm of PBF was significantly associated with variation in all biochemical traits except for the logarithm of apoA-I. FABP2 genotype was significantly associated with variation in the logarithm of TG. PON genotype was significantly associated with variation in the logarithms of total and LDL cholesterol and apoB. APOC3 genotype was significantly associated with variation in the logarithms of TG, HDL cholesterol, apoA-I, TC/HDL ratio, and apoB/A-I ratio. APOE genotype was significantly associated with variation in the logarithms of LDL cholesterol, apoB, TC/HDL ratio, and apoB/A-I ratio.
The basis for the 13 significant phenotype-genotype associations detected by the multivariate ANOVA was then examined with pairwise comparisons. A summary of the significant phenotype-genotype associations from the ANOVA is shown in Table 4⇓. The least-squares means for the ln of TC, TG, LDL cholesterol, HDL cholesterol, apoB, apoA-I, TC/HDL ratio, and apoB/A-I ratio are shown for subjects classified by genotypes of FABP2, PON, APOC3, and APOE. Significant (P<.05) differences in pairwise comparisons between the least-squares mean of a particular genotype and the least-squares means of the other genotypes are indicated by asterisks. For each genotype-phenotype system, the least-squares mean for one homozygous genotype was always significantly different from the least-squares means for the other genotypes.
Table 4⇑ shows that subjects with the T/T genotype of FABP2 codon 54 had significantly higher TG than subjects with the other two genotypes. There was not a significant difference between least-squares mean TG in subjects with the other two FABP2 genotypes (T/A and A/A). This is consistent with either a significant recessive TG-raising effect of the 54T variant of FABP2 or a significant dominant TG-lowering effect of the 54A variant of FABP2.
Table 4⇑ also shows that subjects with the Q/Q genotype of PON codon 192 had significantly lower total and LDL cholesterol and apoB than subjects with the other two PON genotypes. There was not a significant difference between least-squares means of these traits in subjects with the other two genotypes (Q/R and R/R). This is consistent with either a significant recessive apoB-lowering effect of the 192Q variant of PON or with a dominant apoB-raising effect of the 192R variant of PON.
In addition, Table 4⇑ shows that subjects with the T/T genotype of APOC3 position −455 nt had significantly lower TG and TC/HDL ratio and apoB/A-I ratio than subjects with the other two genotypes. Subjects with the T/T genotype of APOC3 position −455 nt also had significantly higher HDL cholesterol and apoA-I than subjects with the other two APOC3 genotypes. There was not a significant difference between least-squares means of these traits in subjects with the other two genotypes (T/C and C/C). This is consistent with either a significant recessive influence of the −455T variant of APOC3 to improve levels of these traits or with a significant dominant influence of the −455C allele to worsen levels of these traits.
Subjects with the E3/2 genotype of APOE had lower least-squares mean ln LDL cholesterol and apoB than subjects with other genotypes (Table 4⇑). Furthermore, there was no difference in the least-squares means of these traits among subjects with the other APOE genotypes. Table 4⇑ also shows that subjects with the E4/4 genotype of APOE had the highest least-squares mean ln TC/HDL ratio and apoB/A-I ratio. Furthermore, there was no difference in the least-squares means of these traits among subjects with the other genotypes. These observations are consistent with a significant codominant influence of the E2 allele toward improving the lipoprotein profile and a significant codominant influence of the E4 allele toward worsening the lipoprotein profile.
The principal finding in this study of young aboriginal Canadians was the identification of several significant associations between candidate genomic variation and variation in fasting plasma concentrations of lipoproteins. We found that 13 of 32 possible genotype-phenotype associations were significant: (1) the FABP2 codon 54 genotype was associated with variation in TG; (2) the PON codon 192 genotype was associated with variation in total and LDL cholesterol and apoB; (3) the APOC3 insulin-response-element genotype was associated with variation in TG, HDL cholesterol, apoA-I, TC/HDL ratio, and apoB/A-I ratio; and (4) the APOE restriction isotype was associated with variation in LDL cholesterol, apoB, TC/HDL ratio, and the apoB/A-I ratio.
Pairwise analyses suggested that the associations resulted from a recessive effect of one allele on the phenotypes. Alternatively, there could have been a dominant effect of the alternate allele for each diallelic system. In either case, the large number of associations identified might have resulted from the relative absence of the confounding influence of age-dependent secondary environmental factors on the phenotypes in this young study sample. These genotype-phenotype associations were thus likely detected both because of a mechanistic impact of the alleles or of a locus in linkage disequilibrium with alleles and because the population was young and relatively free of the influence of time-dependent environmental factors that could have an impact on lipoprotein metabolism. An argument in favor of the effect of age on these associations was the preliminary observation of fewer and less significant associations with plasma lipoproteins when these same markers were studied in adults from the same population (data not shown).
The results from the analysis of association between FABP2 genotype and biochemical traits indicated that the 54T allele was associated with elevated plasma TG concentrations. The FABP2 gene product IFABP is 15 kD in mass and is a member of a family of intracellular lipid-binding proteins. It is expressed in the small intestine,9 suggesting that it plays a role in absorption and intracellular transport of dietary long-chain fatty acids. After absorption, most dietary fatty acids are converted into TG, which leaves the enterocyte in chylomicrons for delivery to peripheral tissues. TG is then hydrolyzed by lipoprotein lipase in the capillaries and the long-chain fatty acids are either locally oxidized or reesterified and returned to the plasma.9 The T54 form of IFABP has increased affinity for long-chain fatty acids compared with the A54 form of IFABP.9 Taken together with our results, this observation is consistent with the hypothesis that the higher plasma TG in subjects with the T54 form of IFABP is related to its increased affinity for and increased intestinal uptake of dietary long-chain fatty acids. Given the small number of T54/T54 homozygotes in our study sample, this association requires confirmation in other samples.
We also found that the low activity variant of paraoxonase, encoded by PON 192Q, was associated with lower plasma total and LDL cholesterol and apoB. Paraoxonase resides on HDL27 and can retard the accumulation of lipid peroxidation products in LDL in vitro.28 Both the serum concentration of paraoxonase and an individual's PON genotype are associated with variation in plasma lipid and lipoprotein concentrations.11 29 Biochemical studies suggested that the low activity variant of paraoxonase had a very low prevalence in native Canadians compared with other populations.30 The high activity variant of paraoxonase, encoded by PON 192R, has been associated with increased susceptibility to CHD.29 30 Paradoxically, the PON 192R allele was also associated with higher plasma HDL cholesterol.29 But in a study of Hutterites, the PON 192Q allele was associated with a lower ratio of LDL:HDL cholesterol,11 which is consistent with our findings in the present study. These apparent disparities may be resolved by considering that PON genetic variation affecting paraoxonase structure may not underlie the genotype-phenotype association. It is possible that, in isolated populations, the PON codon 192 alleles are in linkage disequilibrium with the actual functional determinant of variation in plasma lipoproteins, which may be within a flanking region or within a neighboring gene.
We also found that the presence of the APOC3 genomic variant containing C at position −455 nt of the APOC3 gene was associated with elevated TG and depressed HDL cholesterol compared with the variant containing T at position −455 nt. The basis for this association may have been related to the well-established linkage disequilibrium between the polymorphism at position −455 nt and the alleles of the Sac I polymorphism in the 3′-untranslated region of the APOC3 gene.21 However, a more intriguing possibility is a possible direct mechanistic effect of the DNA change at position −455 nt. In vitro, this variant is resistant to insulin-mediated suppression of APOC3 gene transcription, which was maximally about 50% in the insulin-responsive variant containing T at position −455 nt.18 While mean TG, HDL cholesterol, apoA-I, TC/HDL ratio, and apoB/A-I ratio were within the physiological range in the Sandy Lake sample, they were significantly worse in subjects with at least one C at position −455 nt than in homozygotes for T at position −455 nt. This suggests that the genomic variation of APOC3 that is associated with the loss of insulin-mediated downregulation of in vitro gene expression is also associated with detectable variation in plasma lipoproteins. The findings suggest that the C-containing allele causes slight perturbations of plasma lipoproteins, even in childhood, which in turn might predispose to the development of pathological hypertriglyceridemia given time and the appropriate secondary factors.
We also found that APOE genetic variation was associated with variation in plasma concentrations of apoB in a manner consistent with many other reports from diverse populations.8 Others who have studied the association of APOE genotype on plasma lipoproteins in children showed that while the typical association of APOE alleles with plasma lipids was not apparent in neonates, it was present by age 3.31 Our observations are also consistent with the notion that the differences in plasma traits between subjects with different APOE genotypes are present before adulthood. Our results contrast with results from a study of 133 Greenland Inuit, in which the APOE polymorphism was not found to be associated with plasma lipoproteins.26 Since we studied a larger number of subjects, it is possible that a subtle effect of the APOE genotype could be more readily detected in our sample.
While we observed genotype-phenotype associations detected in an essentially normolipidemic study sample, it must be noted that our sample was relatively large, young, and homogeneous. It is possible that this association might not have been found in other smaller, older, and more heterogeneous samples. The findings also underscore the concept that the effect of an individual genetic variant, even one with significant in vitro functional relevance, is small at the level of complex phenotypes in populations.22 23 A similar phenomenon has been reported for the AGT gene, in which structural changes were associated with modest variation in baseline blood pressure.21 However, the same AGT markers were linked with overt hypertension in kindreds and sibling pairs ascertained on the basis of disease.32 It was proposed that genetic variation of AGT may create a modest increase of blood pressure within the physiological range and that this contributes to susceptibility to hypertension, which becomes manifest in the presence of secondary factors.21 32 This potential impact of a small effect on a baseline phenotype could be analogous to the observed relationship between variation of the candidate genes and the lipoprotein traits that we have studied.
The frequency of FABP2 T54 of 0.142 in the Sandy Lake sample is considerably lower than that observed in Pima Indians9 and whites.9 10 We also found that the low activity variant of paraoxonase, encoded by PON 192Q, had a very low frequency in Sandy Lake compared with almost all other human populations.30 The frequencies of APOE E4 and E2 alleles in the Sandy Lake sample were considerably lower than those reported in white populations.6 The distinctive allele frequencies in Sandy Lake compared with other samples might have resulted from founder effects involving the ancestors of the contemporary community. Archaeologic studies suggested that hunter-gatherers inhabited the Sandy Lake region 6000 years ago.33 The current inhabitants of the Sandy Lake region lived a hunter-gatherer subsistence in northwestern Ontario until about 70 years ago. The present community is largely descended from one clan, and any unusual distribution of these alleles in this small founder group would have affected the allele frequencies seen among the residents of the reserve today.
Alternatively, selection pressure might have produced the present allele frequencies. It is possible that some of the “disease-associated” alleles might have imparted a survival benefit to carriers in the past. For example, heterozygosity for the R192 allele of PON and higher plasma paraoxonase activity could have protected against an environmental or dietary toxin.30 Also, infectious diseases, such as tuberculosis, remain a significant cause of mortality in aboriginal communities. It is possible that mortality due to susceptibility to infectious diseases might have produced the changes in allele frequencies, especially if the alleles studied were in linkage disequilibrium with “resistance” genes in the Sandy Lake sample. Finally, it has been suggested that repeated infection in children can permanently influence the lipoprotein profile,34 and this may represent an important possible gene-environment interaction in which subjects with specific chromosomal haplotypes who survived infection had changes in their lipoprotein phenotypes.
In summary, we have observed that allelic variants of FABP2, PON, APOC3, and APOE genes are associated with variation in plasma lipoprotein traits in young aboriginal subjects from Sandy Lake. The large number of significant but modest associations with plasma lipoprotein traits suggests that a young sample may be useful for detecting such associations because of the relative absence of confounding time-dependent environmental factors. It is possible that even in childhood, genetic variation can produce modest changes in complex traits such as plasma lipoproteins. Such small effects may, however, have clinical significance. For example, a small genetic effect on unstressed baseline variation might be associated with a predisposition to markedly abnormal phenotypes, which become manifest over time when secondary genetic or environmental factors are present. Understanding the impact of environmental factors on a background of genetic predisposition is even more important in native populations, which may develop an increased prevalence of metabolic diseases as their lifestyles westernize.35 36 Furthermore, evaluating the genotype-phenotype associations in young people and following them prospectively may prove to be helpful in defining strategies for CHD risk factor modification in children and adolescents.2
Selected Abbreviations and Acronyms
|AGT||=||gene for angiotensinogen|
|APOC3||=||gene for apoC-III|
|APOE||=||gene for apoE|
|CHD||=||coronary heart disease|
|FABP2||=||gene for intestinal fatty acid–binding protein|
|PBF||=||percent body fat|
|PON||=||gene for serum paraoxonase|
This study was supported by operating grants from the National Institutes of Health (No. 91-DK-01), the Ontario Ministry of Health (No. 04307), the Heart and Stroke Foundation of Ontario (No. T2978), the Medical Research Council of Canada (No. MA-13430), and the St Michael's Hospital Foundation. Dr Hegele is a Career Investigator of the Heart and Stroke Foundation of Ontario. The authors would like to acknowledge the following groups and individuals, whose cooperation was essential in the design and implementation of this project: the chiefs and council of the community of Sandy Lake, the Sandy Lake community surveyors, the Sandy Lake nurses, the staff of the University of Toronto Sioux Lookout program, and the Department of Clinical Epidemiology of the Samuel Lunenfeld Research Institute.
PDAY Research Group. Natural history of aortic and coronary atherosclerotic lesions in youth: findings from the PDAY study. Arterioscler Thromb. 1993;13:1291-1298.
Kwiterovich PO Jr. Dyslipoproteinemia and other risk factors for atherosclerosis in children and adolescents. Atherosclerosis. 1994; 108:S55-S71.
Strong JP. Natural history and risk factors for early human atherogenesis. Pathobiological Determinants of Atherosclerosis in Youth (PDAY) research group. Clin Chem. 1995;41:134-138.
Tracy RE, Newman WP III, Wattigney WA, Berenson GS. Risk factors and atherosclerosis in youth autopsy findings of the Bogalusa Heart Study. Am J Med Sci. 1995;310:S37-S41.
Breslow JL. Genetic basis of lipoprotein disorders. J Clin Invest. 1989;84:373-380.
Mehrabian M, Lusis AJ. Genetic markers for atherosclerosis and related risk factors. In: Lusis AJ, Rotter JI, Sparkes RS, eds. Molecular Genetics of Coronary Artery Disease: Candidate Genes and Processes in Atherosclerosis. Basel, Switzerland: Karger; 1992:363-418.
Baier LJ, Sacchettini JC, Knowler WC, Eads J, Paolisso G, Tataranni PA, Mochizuki H, Bennett PH, Bogardus C, Prochazka M. An amino acid substitution in the human intestinal fatty acid binding protein is associated with increased fatty acid binding, increased fat oxidation and insulin resistance. J Clin Invest. 1995;95:1281-1287.
Mitchell BD, Kammerer CM, O'Connell P, Harrison CR, Manire M, Shipman P, Moyer MP, Stern MP, Frazier ML. Evidence for linkage of postchallenge insulin levels with intestinal fatty acid binding protein (FABP2) in Mexican-Americans. Diabetes. 1995; 44:1046-1053.
Hegele RA, Brunt JH, Connelly PW. A polymorphism of the paraoxonase gene associated with variation in blood pressure in a genetic isolate. Arterioscler Thromb Vasc Biol. 1995;15:89-95.
Serrato M, Marrian AJ. A variant of the human paraoxonase/ arylesterase (HUMPONA) gene is a risk factor for coronary artery disease. J Clin Invest. 1995;96:3005-3008.
Rees A, Stocks J, Sharpe CR, Vella MA, Shoulders CC, Katz J, Jowett NI, Barelle FE, Galton DJ. Deoxyribonucleic acid polymorphism in the apolipoprotein A-I-C-III gene cluster: association with hypertriglyceridemia. J Clin Invest. 1985;76:1090-1095.
Tas S. Strong association of a single nucleotide substitution in the 3′-untranslated region of the apolipoprotein-C-III gene with common hypertriglyceridemia in Arabs. Clin Chem. 1989;35:256-259.
Zeng Q, Dammerman M, Takada Y, Matsunaga A, Breslow JL, Sasaki J. An apolipoprotein CIII marker associated with hypertriglyceridemia in Caucasians also confers increased risk in a west Japanese population. Hum Hered. 1994;95:371-375.
Li W, Dammerman MM, Smith JD, Metzger J, Breslow JL, Leff T. Common genetic variation in the promoter of the human apoCIII gene abolishes regulation by insulin and may contribute to hypertriglyceridemia. J Clin Invest. 1995;96:2601-2605.
Nunez C, Gallagher D, Russell-Aulet M, Heymsfield SB. Bioimpedance analysis of body fat composition: a new measurement approach. Int J Obes Relat Metab Disord. 1994;18:P498.
Kushner RF, Kunigk A, Alspaugh M, Andronis PT, Leitch CA, Schoeller DA. Validation of bioelectrical impedance analysis as a measurement of change in body composition in obesity. Am J Clin Nutr. 1990;52:219-223.
Hegele RA, Brunt JH, Connelly PW. A polymorphism of the angiotensinogen gene is associated with variation in blood pressure in a genetic isolate. Circulation. 1994;90:2207-2212.
Hegele RA, Evans AJ, Tu L, Ip G, Brunt JH, Connelly PW. A gene-gender interaction affecting lipoproteins in a genetic isolate. Arterioscler Thromb. 1994;14:671-678.
Hegele RA, Brunt JH, Connelly PW. Multiple genetic determinants of variation of plasma lipoproteins in a genetic isolate. Arterioscler Thromb Vasc Biol. 1995;15:861-871.
Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with Hha I. J Lipid Res. 1990;31:545-548.
SAS Institute Inc. SAS/STAT User's Guide, Version 6. Cary, NC: SAS Institute Inc; 1990.
de Knijff P, Johansen L, Rosseneu M, Frants RR, Jespersen J, Havekes LM. Lipoprotein profile of a Greenland Inuit population: influence of anthropometric variables, apoE and A4 polymorphism and lifestyle. Arterioscler Thromb. 1992;12:1371-1379.
Blatter M-C, James RW, Messmer S, Barja F, Pometta D. Identification of a distinct human high-density lipoprotein subspecies defined by a lipoprotein-associated protein K-45. Eur J Biochem. 1993;211: 871-879.
Lehtimäki T, Porkka K, Viikari J, Ehnholm C, Akerblom HK, Nikkari T. Apolipoprotein E phenotypes and serum lipids in newborns and 3-year-old children: the Cardiovascular Risk in Young Finns study. Pediatrics. 1994;94:489-493.
Fiddler T, Stevens JR. Killing the Shamen. Toronto, Ontario: Penumbra Press; 1985.
Pesonen E, Rapola J, Viikair J, Turtinen J, Akerblom HK. Altered serum lipid profile after systemic infection in children: risk factor for CHD? Eur Heart J. 1993;14:K7-K11.
Young TK, Moffatt MEK, O'Neil JD. Cardiovascular diseases in a Canadian arctic population. Am J Public Health. 1993;83: 881-887.
Trowell HC, Burkitt DP. Western Diseases: Their Emergence and Prevention. Cambridge, Mass: Harvard University Press; 1981.