Genetic epistasis in the VLDL catabolic pathway is associated with deleterious variations on triglyceridemia in obese subjects

Article metrics



Abdominal obesity and hypertriglyceridemia (the hypertriglyceridemic-waist phenotype) increase cardiovascular risk. The very low-density lipoprotein (VLDL) is a triglyceride (TG)-rich particle. Frequent variations in the genes coding for enzymes and proteins involved in the VLDL catabolism have already been documented. The epistatic effect of such variants on the risk profile associated with abdominal obesity remains to be elucidated.


This study aims to assess the effect of combinations of frequent single-nucleotide polymorphisms (SNPs) in the VLDL catabolic pathway on the relation between abdominal obesity and fasting TG.


Only gene variants in the lipoprotein lipase, apolipoprotein (apo) CIII, hepatic lipase and apo E genes known to be frequent in the general population (allele frequency>5%) were included in this study. The presence of selected SNPs was detected by polymerase chain reaction-restriction fragment length polymorphisn in a sample of 640 non-diabetic French Canadians at high cardiovascular risk (405 obese, 235 non-obese).


Carrying more than two frequent gene variants involved in the VLDL catabolic pathway significantly increased the risk of hyperTG (odds ratio of TG>1.7 mmol/l=4.15; P=0.001). This effect was proportional to the number of SNPs and genes involved and was significantly amplified by the presence of abdominal obesity defined on the basis of waist circumference.


When combined with abdominal obesity, epistasis in the VLDL pathway has a deleterious effect on fasting TG and coronary artery disease risk profile according to the TG threshold (1.7 mmol/l) used in medical guidelines for the assessment of the metabolic syndrome and associated risk.


Several markers are available to assess coronary artery disease (CAD) risk. In particular, an atherogenic triad combining hyperapolipoprotein B (hyperapo B), small and dense low-density lipoproteins (LDL), and hyperinsulinemia was found to be effective in identifying high-risk CAD patients.1 However, measurement of these variables is expensive and not always available to clinicians. In this regard, it has recently been suggested that the ‘hypertriglyceridemic-waist’ phenotype, in which subjects are characterized by a waist circumference over 90 cm in men and over 85 cm in women combined with hypertriglyceridemia (hyperTG), was a simple and easily available correlate of the metabolic atherogenic triad and CAD risk.2, 3, 4 Several pathways are involved in the expression of the atherogenic triad and may contribute to the CAD risk. This is the case for the triglyceride (TG)-rich very low-density lipoprotein (VLDL) catabolic pathway. The catabolic cascade of the VLDL particle sequentially involves several enzymes, regulatory elements and apolipoproteins (apo), each coded by a specific gene (Figure 1). In many of these genes, several single-nucleotide polymorphisms (SNPs) have already been documented, some of them having been reported as frequent (>5%) in the general population, thus increasing the odds of simultaneously carrying two or more of these SNPs. As the independence of common gene variants transmission is most often associated with independence of the effect, this is not always relevant from a clinical or public health standpoint. However, in a sequential metabolic cascade such as the VLDL catabolic pathway, the combined effect of gene variations acting in the same direction can be synergistic on the risk profile. Furthermore, in a context of independence of transmission, such combinations can be frequent in the general population, contributing to the clinical and public health utility of such information.

Figure 1

Sites of action of the studied genotypes in the VLDL catabolic pathway. PPARα=peroxisome proliferator-activated receptor α; LPL=lipoprotein lipase; Apo E=apolipoprotein E; Apo B=apolipoprotein B; HL=hepatic lipase; Apo CII-CIII=apolipoproteins CII and CIII; VLDL=very low-density lipoprotein; IDL=intermediate-density lipoprotein; LDL=low-density lipoprotein.

The aim of this study was thus to estimate the effect of different combinations of gene variants involved in the VLDL metabolism on the expression of the ‘hypertriglyceridemic-waist’ phenotype in a sample of 640 non-diabetic French Canadians. Specifically, we have assessed fasting TG concentrations associated with combinations of frequent SNPs in genes involved in the VLDL catabolic cascade, in relation to the level of abdominal obesity (defined by a waist circumference 90 cm in men and 85 cm in women). Gene variants were selected on the basis of their allele frequency (>5%) in the general population, as estimated in a random sample of the Cardiovascular Study of Santé Québec (1991). The selected variants were lipoprotein lipase (LPL) gene variants (D9N, N291S and P207L), the apo CIII-SstI, the hepatic lipase (HL)-T514 allele (HL-514 C/T), the peroxisome proliferator-activated receptor (PPAR)α-V162 allele (PPARα-L162 V), the PPARγ-A12 allele (PPARγ-P12A) and the apo e2 and apo e4 alleles. The analyses showed that, individually, LPL variants (D9N, N291S and P207L), the apo CIII-SstI, the HL-T514 allele and the apo e2 allele were associated with a significant elevation of plasma TG concentrations. The combination of two or more of these hyperTG variants had an incremental effect on plasma TG. The influence of these gene variants and their combination on plasma TG was significantly exacerbated by the presence of abdominal obesity, defined on the basis of waist circumference.


Subjects and clinical data

This study comprised a sample of 640 French-Canadian untreated subjects (354 men and 286 women) from the Saguenay-Lac-Saint-Jean region (Quebec, Canada), selected on the basis of having a positive family history of dyslipidemia, CAD or type 2 diabetes. All subjects were followed at the Chicoutimi Hospital Lipid Clinic and had agreed to participate in studies on genetic determinants of type 2 diabetes or CAD.5, 6 Subjects homozygous for a null LPL gene mutation, those with familial hypercholesterolemia (FH), the apo e2/e4 genotype, or diagnosed with type 2 diabetes, as well as those taking drugs known to affect blood lipid levels were excluded. Type 2 diabetes was defined according to the World Health Organization criteria as a 2-h glucose concentration >11.1 mmol/l following a 75 g oral glucose load, whereas a normal glucose tolerance state was characterized as a 2-h glucose concentration below 7.8 mmol/l. The presence of CAD was determined based on clinical history of retrosternal pain, electrocardiogram and clinically documented myocardial infarction or angiographic evidence of coronary lesions, as described previously.7, 8 The prevalence of SNPs in the general population was evaluated among a random sample of 1842 subjects from the Cardiovascular Study of Santé Québec (1991).

Blood samples were obtained after a 12h-overnight fast from the antecubital vein into vacutainer tubes containing ethylenediaminetetraacetic acid. The high-density lipoprotein (HDL) subfraction was obtained after precipitation of the LDL (d>1.006 g/ml) in the infranatant with heparin and MnCl2.9 Cholesterol and TG levels were enzymatically measured on a Multiparity Analyzer CX7 (Beckman). Apo B levels were determined using nephelometry. Subjects gave informed consent to participate in this study and were assigned a code systematically denominalizing all clinical data.10 This project has received the approval of the Chicoutimi Hospital Ethics Committee.


The PPARα-L162 V and PPARγ-P12A variants, the presence of P207L, G188E, D9N and N291S variants in the LPL gene and the apo E genotype were identified by a mismatch polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP)-based method, as described previously.5, 11, 12, 13, 14, 15, 16 The presence of apo CIII-SstI and HL -514 C/T SNPs was identified using the fluorescence polarization detection method.17 Molecular screening of FH included the detection of nine LDL-receptor gene mutations explaining the majority of cases in the province of Quebec. Two deletions (5 kb and >15 kb) were detected by Southern blotting.18 The W66G, E207 K, C646Y, C152W, R329X, C347R and Y468X mutations were detected by dot-blot hybridization of genomic DNA amplified by PCR with allele-specific oligonucleotide probes or by PCR-based restriction fragment analysis.19, 20, 21

Statistical analysis

Categorical variables were compared using the Pearson χ2-statistic, whereas group differences for continuous variables were compared with the Student's unpaired two-tailed t-test or an analysis of variance (ANOVA), followed by the Bonferonni post hoc test. Multivariate logistic regression models were built in order to calculate the relative odds of exhibiting plasma TG concentrations above the upper standard limit of 1.7 mmol/l, as defined by several clinical guidelines,22, 23 associated with the number of carried hyperTG genotypes (namely the apo CIII-SstI, the HL −514 T allele and the apo E2/2 genotype or the apo E2/3 genotype in the presence of an additional primary or secondary dyslipidemic factor), according to waist circumference. The likelihood ratio statistic was used to compare regression models. Odds ratio (OR) are reported. The effects of discrete variables were evaluated dichotomously, according to their absence or presence in subjects. Univariate linear models were also constructed, and the significance of interaction between waist circumference groups and the number of hyperTG genotypes in the resulting models was assessed with an F test. All statistical analyses were performed with the SPSS package (release 11.0, SPSS, Chicago, IL, USA).


The distribution of all hyperTG genotypes was in Hardy–Weinberg equilibrium and was not affected by gender, except for the HL-514 T allele, which was more frequent among men in this sample. As shown in Table 1, the apo E4, PPARγ-A12 and PPARα-V162 alleles did not significantly affect plasma TG concentrations, whereas the apo CIII-SstI, HL-514 T, apo e2 alleles and all selected LPL variants were associated with increased plasma TG concentrations compared to non-carriers. According to allele frequencies in the general population, >12% of French Canadians are likely to carry two of the selected hyperTG genotypes, whereas 1.3% are potential carriers of three or more of these variants (Table 2).

Table 1 Prevalence of frequent VLDL catabolic pathway-related genotypes in the general population and signification levels of their effects on plasma triglycerides and apolipoprotein B concentration increase
Table 2 Population AF of hyperTG genotype combinations involved in the VLDL catabolic pathway

Table 3 presents subjects' characteristics in relation to the number of hyperTG genotypes carried. As shown, plasma TG levels tended to increase with the number of hyperTG SNPs involved. In the group of subjects carrying more than two hyperTG genotypes, the mean plasma TG concentrations was increased by 2-fold (4.90±0.54 mmol/l as compared with 2.11±0.13 mmol/l; P<0.05), and HDL-cholesterol concentrations were significantly lower compared to non-carriers. Figure 2 shows that in the presence of abdominal obesity, combining three or more hyperTG genotypes was associated with an increased risk to exhibit fasting plasma TG concentrations above the threshold (1.7 mmol/l) suggested in clinical guidelines for the diagnosis of the metabolic syndrome and CAD risk assessment (OR=4.15; P=0.001). The OR remained comparable when carriers of three or four hyperTG genotypes were analyzed separately (data not shown), and thus the two groups were combined. Among carriers of two genotypes, the risk of hyperTG tended to increase only in the presence of a waist circumference 90 cm in men or 85 cm in women (OR=1.73; P=0.05). The risk of hyperTG remained significant when controlling for the type of hyperTG genotype, and the results were independent of the nature of the gene combination considered. However, Table 4 illustrates that although abdominal obesity increases the odds of hyperTG, the distribution of the different combinations of genotypes was not significantly different between the groups of obese vs. non-obese subjects.

Table 3 Subjects' characteristics according to the number of frequent hyperTG genotypes carried
Figure 2

Estimated risk (odds ratio) to exhibit elevated plasma TG levels (>1.7 mmol/l) associated with the presence of frequent combinations of hyperTG genotypes, after stratification for waist circumference. Odds ratios were obtained from multivariate logistic regression analyses taking into account the effect of other TG metabolism confounders, such as age, gender and the presence of coronary artery disease. Subjects with type 2 diabetes or taking medication known to affect plasma TG were excluded from the present study. HyperTG genotypes considered are listed in Table 2. NS=P>0.1.

Table 4 Distribution of combinations of selected genotypes significantly affecting plasma TG concentrations, according to waist circumference

The effect of abdominal obesity on the odds of hyperTG (TG>1.7 mmol/l) was also tested separately on each hyperTG genotype (Figure 3). Abdominal obesity was independently associated with an increased plasma TG concentration in LPL, apo E2/2 or apo CIII allele carriers. The independent association of the apo E2 allele with plasma TG concentrations was more complex to assess. Indeed, lean individuals being heterozygous for the apo E2 allele and not carrying another hyperTG genotype tended to have a normal lipid profile, and the odds of finding TG concentrations above 1.7 mmol/l in this group was 0.47 (P=0.036). In comparison, abdominally obese apo E2 allele carriers, particularly E2 homozygotes or E2 hererozygotes simultaneously carrying another hyperTG genotype, tended to exhibit hyperTG (OR=2.10; P=0.07). Figure 4 shows a significant interaction between the number of carried hyperTG genotypes and the presence of a waist girth over 90 cm in men or over 85 cm in women (P=0.011). The effect on plasma TG tended to be stronger as the number of genotypes increased, especially in abdominally obese subjects.

Figure 3

Estimated risk of hyper TG (OR of plasma TG levels >1.7 mmol/l) associated with frequent hyperTG genotypes considered at the gene level, after stratification for waist circumference. Odds ratios were obtained from multivariate logistic regression analyses taking into account the effect of TG metabolism confounders, such as age, gender and the presence of coronary artery disease. The hyperTG genotypes presented are LPL D9N, N291S and P207L variants (LPL), the apo E2 allele in the absence of abdominal obesity (E2 a) or in the presence of abdominal obesity and additional primary or secondary dyslipidemic factor (E2b), the subgroup of individuals with waist circumference 90 cm for men or 85 cm for women being homozygotes for the apo E2 allele (E2/2), the HL-T514 allele (HL) and the apo CIII-SstI polymorphism (apo CIII). NS refers to a P value >0.1.

Figure 4

Plasma TG concentrations (means±s.e.m.) according to the presence of frequent combinations of hyperTG genotypes (identified in Table 2), after stratification for waist circumference. (a) Significantly different (P<0.05) as compared to levels measured in groups with no hyperTG genotype (b) with one hyperTG genotype and (c) with two hyperTG genotypes. *P value corresponds to the significance of interaction between the number of hyperTG genotypes carried and waist circumference group. **Corrected for the effect of subjects' age.


The ‘hypertriglyceridemic-waist’ phenotype was proposed as a crude marker of a proatherogenic triad (hyperinsulinemia, hyperapo B and dense LDL).2 This simple clinical phenotype was found to be associated with a 80% probability of bearing the atherogenic metabolic triad features and therefore presenting a higher risk for CAD.2, 3 We have provided further evidence of a significant contribution of the ‘hypertriglyceridemic-waist’ phenotype in the modulation of the CAD risk generally associated with hyperglycemia.4 Several genes have the potential to affect this phenotype. The clinical and public health utility and validity of such genetic determinants remains to be determined. Part of the challenge resides in identifying gene variants (and combination of variants) that may have modest effects on the ‘hypertriglyceridemic-waist’ relative risk (RR), but bear significant population attributable risks (PAR).

Different strategies have been suggested for the identification of multiple loci involved in complex traits. One of them resides in isolating a set of single gene variants, under liberal statistical criteria, and then evaluating possible interactions among them. This method has proven to be more powerful than traditional analyses, even for studies dealing with hundreds of loci.24 The analytic strategy of the present study, however, is more linear. The approach used to identify gene variants and gene combinations associated with the ‘hypertriglyceridemic-waist’ phenotype was to target a cascade of events physiologically connected within a sequential cascade, namely the VLDL catabolic pathway. The VLDL particle is a TG-rich apo B-containing lipoprotein. Its catabolism leads to the formation of intermediate-density lipoprotein (IDL) and LDL particles, both associated with a gradient of atherogenicity, which depends on the number and quality (including the density) of these particles. Defects within the VLDL catabolism are well documented and are frequently observed in the presence of abdominal obesity, type 2 diabetes, insulin resistance or CAD. Several enzymes, apolipoproteins, transfer proteins, transducers, regulatory elements and receptors are sequentially involved in the VLDL pathway. Most of the genes coding for these enzymes and proteins have already been identified. Several mutations and polymorphisms have been identified in these genes, some of them being known to interfere with protein production or activity, and to increase plasma apo B or TG concentrations and CAD risk. The allele frequency of these gene variants varies across populations.

The gene variants in the VLDL catabolic pathway selected for this study were reported at an allele frequency (AF)>5% in different populations,25, 26, 27, 28, 29 including French Canadians, so that the probability of exhibiting simultaneously two or more of these variants was high enough (>6%) to observe and estimate the effect of gene combinations. These gene variants are also known to interfere with the VLDL pathway and contribute to variations in either plasma apo B or TG concentration, type 2 diabetes or CAD risk.25, 26, 27 These variants were LPL gene variants (D9N, N291S and P207L), the apo CIII-SstI SNP, the HL-T514 allele (HL-514 C/T), the PPARα-V162 allele (PPARα-L162 V), the PPARγ-A12 allele (PPARγ-P12A) and the apo e2 and apo e4 alleles. The AF values of all the gene variants listed above has been assessed in random samples of the Quebec population.5, 25, 28, 29 In other populations, the AF of less common alleles for PPARα-V162, apo CIII-SstI and HL–514 C/T SNPs have been estimated at 6.3, 8.6 and 19.3%, respectively, in the Framingham offspring's study, whereas the prevalence of the e2 allele has been reported as high as 10% in several populations.25, 26, 27

The selected SNPs were presumed to affect the activity of enzymes, transducers or proteins involved in the VLDL catabolism in the same direction (Figure 1). Interference with the activity of all or any of these enzymes or proteins has the potential to delay TG hydrolysis (LPL, HL), decrease LPL availability for hydrolysis (Apo CIII) or generate interference with receptor recognition by TG-rich VLDL or IDL (apo E2);25, 26, 27, 28 the most frequent clinical consequence is that the plasma apo B or TG concentration might increase as well as the CAD risk. A recent study has observed that the apo CIII-SstI SNP increases fasting TG levels among carriers of the LPL P207L mutation.30 Reilly et al.31 have also recently shown that subjects who were heterozygotes for four SNPs in three lipase genes (LPL, HL and endothelial lipase) had TG levels beyond the effect of individual lipase SNPs, suggesting a synergistic association. These studies highlight the importance of studying genes in combination.

Our results suggest that the combination of the effect of two or more frequent gene variants involved in the VLDL catabolic pathway has an incremental and synergistic effect on plasma TG, especially in a context where the workload is likely to be increased (as frequently observed in presence of abdominal obesity). In the presence of abdominal obesity, the gene combinations we have studied were associated with a significant increase in fasting plasma TG concentrations above the threshold suggested in different guidelines for the diagnosis of the metabolic syndrome and CAD risk assessment (1.7 mmol/l).22, 23 This trend remained unchanged when controlling for the type of hyperTG genotype, and the results were independent of the nature of the gene combination considered. Thus, from a clinical standpoint, the most significant difference between the different gene combinations studied is not likely to be their specific effect on plasma TG, which appears important but comparable, but may rather reside in other dimensions, including their effect on drug response to lipid-lowering therapy. Indeed, combinations involving the apo E2 allele might be associated with a significantly better and dosage-specific response to both fibrates and statins, as previously suggested by our group and others.32, 33

The RR of hyperTG associated with each of these frequent genotypes could be modest when considered at the unit level, but a modest effect on the RR could translate into a significant PAR and contribute to a significant proportion of the variance of plasma TG concentrations in the general population.34 It is obvious that the allele frequency of the selected SNPs will vary from one population to another, thus modifying the relative contribution of the different gene combinations to the PAR across populations. All subjects in our sample were selected from a cohort originally created for the study of genetic determinants of type 2 diabetes or CAD. Considering that the genotypes studied in the present research are potential modulators of these two phenotypes,25, 26, 27, 28 we did not expect their allele frequencies in our sample to be comparable with those found in the general population. However, as the biological impact of gene variants appears consistent in magnitude and direction across continents and populations,35 our sample had the advantage of providing us with a sufficient number of SNP carriers in each group to estimate the risk of hyperTG associated with different SNP combinations. In order to assess their utility and validity, the results presented herein should however be replicated in other studies and in diversified populations.

The contribution of frequent alleles affecting the VLDL pathway to the PAR of hyperTG should also be considered in the context of the endemization of obesity worldwide.36 The present study suggests that abdominal obesity is an important factor worsening the genotypes' effects. This is concordant with our earlier studies.37, 38, 39, 40 Our results specifically illustrate that combinations of genetic factors sequentially involved in the VLDL catabolic pathway might influence the expression of the ‘hypertriglyceridemic-waist’ phenotype. Prevention of the risk associated with the ‘hypertriglyceridemic-waist’ phenotype importantly relies on intervention on life habits (diet, physical activity, etc.) and modifiable risk factors.41 However, the understanding of the influence of genetic factors (including epistasis) on the expression of this phenotype, or any other element of the metabolic syndrome and CAD risk, constitutes a new level of information available to public health authorities and health professionals. In this regard, genetic health determinants should be considered as one of the key components within a global systems approach needed to better understand the modulators of the risk of disease at the population or community level. Considering the endemization of obesity and related disorders worldwide,36 it is also of interest to document genetic modulators of the ‘hypertriglyceridemic-waist’ phenotype with the intent of identifying new potential targets for treatment and prevention.


  1. 1

    Lamarche B, Tchernof A, Mauriege P, Cantin B, Dagenais GR, Lupien PJ et al. Fasting insulin and apolipoprotein B levels and low-density lipoprotein particle size as risk factors for ischemic heart disease. JAMA 1998; 279: 1955–1961.

  2. 2

    Lemieux I, Pascot A, Couillard C, Lamarche B, Tchernof A, Almeras N et al. Hypertriglyceridemic waist: A marker of the atherogenic metabolic triad (hyperinsulinemia; hyperapolipoprotein B; small, dense LDL) in men? Circulation 2000; 102: 179–184.

  3. 3

    St-Pierre J, Lemieux I, Vohl MC, Perron P, Tremblay G, Despres JP et al. Contribution of abdominal obesity and hypertriglyceridemia to impaired fasting glucose and coronary artery disease. Am J Cardiol 2002; 90: 15–18.

  4. 4

    St-Pierre J, Lemieux I, Perron P, Brisson D, Santure M, Vohl MC et al. Relation of the hypertriglyceridemic waist phenotype to earlier manifestations of coronary artery disease in patients with glucoe intolerance and type 2 diabetes mellitus. Am J Cardiol 2007; 99: 369–373.

  5. 5

    Vohl MC, Lepage P, Gaudet D, Brewer CG, Betard C, Perron P et al. Molecular scanning of the human PPARα gene: association of the L162 V mutation with hyperapobetalipoproteinemia. J Lipid Research 2000; 41: 945–952.

  6. 6

    Gaudet D, Arsenault S, Perusse L, Vohl MC, St-Pierre J, Bergeron J et al. Glycerol as a correlate of impaired glucose tolerance: dissection of a complex system by use of a simple genetic trait. Am J Hum Genet 2000; 66: 1558–1568.

  7. 7

    Gaudet D, Vohl MC, Perron P, Tremblay G, Gagne C, Lesiege D et al. Relationships of abdominal obesity and hyperinsulinemia to angiographically assessed coronary artery disease in men with known mutations in the LDL receptor gene. Circulation 1998; 97: 871–877.

  8. 8

    Gaudet D, Vohl MC, Julien P, Tremblay G, Perron P, Gagne C et al. Relative contribution of low-density lipoprotein receptor and lipoprotein lipase gene mutations to angiographically assessed coronary artery disease among French Canadians. Am J Cardiol 1998; 82: 299–305.

  9. 9

    Havel RJ, Eder HA, Bragdon JH . The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest 1955; 34: 1345–1354.

  10. 10

    Gaudet D, Arsenault S, Bélanger C, Hudson TJ, Perron P, Bernard M et al. Procedure to protect confidentiality of familial data in community genetics and genomic research. Clin Genet 1999; 55: 259–264.

  11. 11

    Bijvoet SM, Hayden MR . Mismatch PCR: a rapid method to screen for the Pro207 → Leu mutation in the lipoprotein lipase (LPL) gene. Hum Mol Genet 1992; 1: 541.

  12. 12

    Monsalve MV, Henderson H, Roederer G, Julien P, Deeb S, Kastelein JJ et al. A missense mutation at codon 188 of the human lipoprotein lipase gene is a frequent cause of lipoprotein lipase deficiency in persons of different ancestries. J Clin Invest 1990; 86: 728–734.

  13. 13

    De Bruin TW, Mailly F, van Barlingen HH, Fisher R, Castro Cabezas M, Talmud P et al. Lipoprotein lipase gene mutations D9N and N291S in four pedigrees with familial combined hyperlipidaemia. Eur J Clin Invest 1996; 26: 631–639.

  14. 14

    Ma Y, Wilson BI, Bijvoet S, Henderson HE, Cramb E, Roederer G et al. A missense mutation (Asp250Asn) in exon 6 of the human lipoprotein lipase gene causes chylomicronemia in patients of different ancestries. Genomics 1992; 13: 649–653.

  15. 15

    Hixson JE, Vernier DT . Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990; 31: 545–548.

  16. 16

    Yen CJ, Beamer BA, Negri C, Silver K, Brown KA, Yarnall DP et al. Molecular scanning of the human peroxisome proliferator activated receptor gamma (hPPAR gamma) gene in diabetic Caucasians: identification of a Pro12Ala PPAR gamma 2 missense mutation. Biochem Biophys Res Commun 1997; 241: 270–274.

  17. 17

    Kwok PY . SNP genotyping with fluorescence polarization detection. Hum Mut 2002; 19: 315–323.

  18. 18

    Ma YH, Betard C, Roy M, Davignon J, Kessling AM . Identification of a second French Canadian LDL receptor gene deletion and development of a rapid method to detect both deletions. Clin Genet 1989; 36: 219–228.

  19. 19

    Leitersdorf E, Tobin EJ, Davignon J, Hobbs HH . Common low-density lipoprotein receptor mutations in the French Canadians population. J Clin Invest 1990; 85: 1014–1023.

  20. 20

    Vohl MC, Couture P, Moorjani S, Torres AL, Gagne C, Despres JP et al. Rapid restriction fragment analysis for screening four point mutations of the low-density lipoprotein receptor gene in French Canadians. Hum Mutat 1995; 6: 243–246.

  21. 21

    Couture P, Vohl MC, Gagne C, Gaudet D, Torres AL, Lupien PJ et al. Identification of three mutations in the low-density lipoprotein receptor gene causing familial hypercholesterolemia among French Canadians. Hum Mutat 1998; (Supp 1): S226–S231.

  22. 22

    Genest J, Frohlich J, Fodor G, McPherson R . Working Group on Hypercholesterolemia and Other Dyslipidemias. Recommendations for the management of dyslipidemia and the prevention of cardiovascular disease: summary of the 2003 update. CMAJ 2003; 169: 921–924.

  23. 23

    Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497.

  24. 24

    Marchini J, Donnelly P, Cardon LR . Genome-wide strategies for detecting multiple loci that influence complex diseases. Nat Genet 2005; 37: 413–417.

  25. 25

    Mahley RW, Rall SC . Type III hyperlipoproteinemia (dysbetalipoproteinemia): The role of apolipoprotein E in normal and abnormal lipoprotein metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds). The metabolic basis of inherited disease 8th ed. McGraw-Hill Book Co: New York, 2000. pp 2835–2862.

  26. 26

    Russo GT, Meigs JB, Cupples LA, Demissie S, Otvos JD, Wilson PW et al. 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.

  27. 27

    Couture P, Otvos JD, Cupples LA, Lahoz C, Wilson PW, Schaefer EJ et al. Association of the C-514T polymorphism in the hepatic lipase gene with variations in lipoprotein subclass profiles: The Framingham Offspring Study. Arterioscler Thromb Vasc Biol 2000; 20: 815–822.

  28. 28

    Murthy V, Julien P, Gagne C . Molecular pathobiology of the human lipoprotein lipase gene. Pharmacol Ther 1996; 70: 101–135.

  29. 29

    Garenc C, Aubert S, Laroche J, Rousseau F, Bergeron J, Julien P . Determination of the apoC3 C-482 T and SstI allele frequencies in the population of Québec City. Can J Cardiol 2001; 17: 192C.

  30. 30

    Garenc C, Couillard C, Laflamme N, Cadelis F, Gagne C, Couture P et al. Effect of the APOC3 Sst I SNP on fasting triglyceride levels in men heterozygous for the LPL P207L deficiency. Eur J Hum Genet 2005; 13: 1159–1165.

  31. 31

    Reilly MP, Foulkes AS, Wolfe ML, Rader DJ . Higher order lipase gene association with plasma triglycerides. J Lipid Res 2005; 46: 1914–1922.

  32. 32

    Brisson D, Ledoux K, Bosse Y, St-Pierre J, Julien P, Perron P et al. Effect of apolipoprotein E, peroxisome proliferator-activated receptor alpha and lipoprotein lipase gene mutations on the ability of fenofibrate to improve lipid profiles and reach clinical guideline targets among hypertriglyceridemic patients. Pharmacogenetics 2002; 12: 313–320.

  33. 33

    Nestel P, Simons L, Barter P, Clifton P, Colquhoun D, Hamilton-Craig I et al. A comparative study of the efficacy of simvastatin and gemfibrozil in combined hyperlipoproteinemia: prediction of response by baseline lipids, apo E genotype, lipoprotein(a) and insulin. Atherosclerosis 1997; 129: 231–239.

  34. 34

    Hennekens CH, Buring JE . Epidemiology in Medicine. Little Brown and co.: Boston, 1987.

  35. 35

    Ioannidis JP, Ntzani EE, Trikalinos TA . ‘Racial’ differences in genetic effects for complex diseases. Nat Genet 2004; 36: 1312–1318.

  36. 36

    Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM . Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA 2004; 291: 2847–2850.

  37. 37

    St-Pierre J, Miller-Felix I, Paradis ME, Bergeron J, Lamarche B, Despres JP et al. Visceral obesity attenuates the effect of the hepatic lipase -514C&gt;T polymorphism on plasma HDL-cholesterol levels in French-Canadian men. Mol Genet Metab 2003; 78: 31–36.

  38. 38

    St-Pierre J, Lemieux I, Miller-Felix I, Prud'homme D, Bergeron J, Gaudet D et al. Visceral obesity and hyperinsulinemia modulate the impact of the microsomal triglyceride transfer protein -493G/T polymorphism on plasma lipoprotein levels in men. Atherosclerosis 2002; 160: 317–324.

  39. 39

    Vohl MC, Lamarche B, Moorjani S, Prud'homme D, Nadeau A, Bouchard C et al. The lipoprotein lipase HindIII polymorphism modulates plasma triglyceride levels in visceral obesity. Arterioscler Thromb Vasc Biol 1995; 15: 714–720.

  40. 40

    Vohl MC, Lamarche B, Pascot A, Leroux G, Prud'homme D, Bouchard C et al. Contribution of the cholesteryl ester transfer protein gene TaqIB polymorphism to the reduced plasma HDL-cholesterol levels found in abdominal obese men with the features of the insulin resistance syndrome. Int J Obes Relat Metab Disord 1999; 23: 918–925.

  41. 41

    Poirier P, Despres JP . Exercise in weight management of obesity. Cardiol Clin 2001; 19: 459–470.

Download references


During this research, D Brisson was the recipient of a doctoral industry-partnered studentship from the Canadian Institutes for Health and Research (CIHR), in collaboration with Fournier Pharma, and from the ‘Réseau en santé cardiovasculaire (RSCV)- Fonds de la recherche en santé du Québec (FRSQ)’. J St-Pierre is the recipient of the ‘Walter & Jessie Boyd & Charles Scriver’ MD/PhD Studentship Award from the CIHR, the Canadian Genetic Disease Network, the Canadian Gene Cure Foundation and Theratechnologies (A Jean Degranpré Scholarship Award). JP Després is Scientific Director of the International Chair on Cardiometabolic Risk, which is supported by an unrestricted grant from Sanofi Aventis awarded to Université Laval. D Gaudet is the Canada Research Chair in preventive genetics and community genomics ( This project was supported by the ECOGENE-21 project from the CAHR/CIHR program (grant #CAR43283) and AstraZeneca Canada Inc.

Author information

Correspondence to D Gaudet.

Rights and permissions

Reprints and Permissions

About this article


  • hypertriglyceridemia
  • waist circumference
  • genetics
  • epistasis
  • risk factor

Further reading