Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Ann Epidemiol. Author manuscript; available in PMC 2009 November 1.
Published in final edited form as:
PMCID: PMC2582040

Longitudinal Changes in Triglycerides According to ANGPTL4[E40K] Genotype and Longitudinal Body Weight Change in the Atherosclerosis Risk in Communities Study



Allelic variation in the adipokine angiopoietin-like 4 gene (ANGPTL4[E40 K]) has been cross-sectionally associated with triglycerides, but the effects of genotype, or the interaction between genotype and body weight, on longitudinal triglyceride change have not been studied.


Body weight, triglycerides, and ANGPTL4[E40 K] genotype were determined at baseline (1987–1989) and at 3 follow-up exams (1990–1992, 1993–1995, 1996–1998) in 7,939 white ANGPTL4 [E40 K] G allele homozygotes and 344 A allele carriers. Changes in body weight and triglycerides were characterized as the difference between exam 4 and baseline measurements.


ANGPTL4[E40 K] A allele carriers had triglyceride concentrations that were 15 to 18 mg/dL lower than G allele homozygotes at all exams (P < 0.0001). Triglycerides increased in both genotype groups over the 9-year period (+19.1 ± 0.9 and +16.2 ± 4.1 mg/dL in GG and GA/AA, respectively; P difference, 0.48). Weight gain was associated with increased triglycerides to a comparable degree in both genotype groups (+5.5 ± 0.3 and +4.3 ± 0.9 mg/dL per 2-kg increase in body weight in GG and GA/AA, respectively, p interaction = 0.30).


Differences in triglyceride concentrations between ANGPTL4[E40 K] A allele carriers and G allele homozygotes are maintained over time, but the degree of increase in triglycerides was similar between groups and was not modified by body weight changes.

Keywords: ANGPTL4[E40 K], Triglycerides, Body Weight, Atherosclerosis Risk in Communities (ARIC) study, Longitudinal Study


The adipokine angiopoietin-like 4 (ANGPTL4) is thought to regulate fatty acid transport among tissues by inhibiting lipoprotein lipase, a key enzyme in triglyceride (TG) metabolism (13). Consistent with its suspected role, a recent cross-sectional analysis in white participants from three different cohorts showed a strong association between baseline TG levels and the non-synonymous sequence variant [E40 K] in the ANGPTL4 gene (4). Homozygous (AA) and heterozygous A allele carriers (GA) had significantly lower TG concentrations compared with their homozygous wild-type counterparts (GG). The influence of this polymorphism on longitudinal changes in TG has not yet been investigated. Given the relation between TG concentrations and the risk of cardiovascular disease (CVD), the effect of ANGPTL4[E40 K] polymorphisms may have important implications for risk of incident disease, especially if one considers the potential for lifetime low TG concentrations to have a greater impact than might be expected from a single measure of TG (analogous to that reported for low-density lipoprotein cholesterol and PCSK9 polymorphisms) (5).

Changes in body weight predict changes in TGs (69), but it is possible that the influence of body weight change on changes in TG concentration may be dampened in individuals with a genetic predisposition to lower TG concentrations (i.e., ANGPTL4[E40 K] A allele carriers). Therefore we characterized the 9-year longitudinal changes in TG in ANGPTL4[E40 K] GG and A allele carriers. We also examined the potential interaction between longitudinal body weight change and ANGPTL4[E40 K] genotype with respect to longitudinal TG changes. These analyses were conducted with data from white men and women who participated in the Atherosclerosis Risk in Communities (ARIC) study, one of the three cohorts involved in the original cross-sectional study demonstrating the strong effect of ANGPTL4[E40 K] variation and TG concentration (4).



The ARIC study was initiated in 1987 to investigate the etiology and occurrence of atherosclerosis and atherosclerotic diseases (10). Men and women, 45 to 64 years of age, were recruited from four U.S. communities: Forsyth County, North Carolina; Jackson, Mississippi (African Americans only); northwest Minneapolis suburbs, Minnesota; Washington County, Maryland. After providing informed consent, 15,792 participants were enrolled in the study (8,710 women and 7,082 men). Study protocols were approved by the institutional review boards at each center. Because of limited variation among non-whites in ANGPTL4 at the locus studied, non-white participants (n = 4,314) were excluded from the current investigation. The current investigation also excluded participants who had a history of diabetes (n = 1,046), who were taking cholesterol-lowering medications (n = 386), who were not typed for ANGPTL4[E40 K] (n = 973), who fasted less than 8 hours (n = 234), or who regularly consumed large quantities of alcohol (>20 g/d for women and >30 g/d for men) (n = 840). Participants missing baseline measures of TG concentration or with no follow-up measures of TG concentration were also excluded (n = 401). After these exclusions, 3,708 white men and 4,575 white women remained for analysis.

TG Measurement and Anthropometry Measurements

Plasma TGs were measured by enzymatic procedures (11), with reagents from Boehringer Mannheim Biochemical (analysis adapted for use in Cobas-Bio Analyzer, Roche). External control consisted of successful participation in the Lipid Standardization Program of the Centers for Disease Control and Prevention. The coefficient of variation for TGs was 7% (12). Changes in TGs were calculated as the difference between baseline and exam 4 measurements.

Body mass index (BMI) was calculated from measured weight (in kilograms)/height (in meters).2 Waist circumference was measured to the nearest centimeter. Change in body weight and waist circumference were calculated as the difference between baseline and exam 4 measurements.


The ANGPTL4[E40 K] polymorphism was measured as previously described (4) using isolated genomic DNA and the TaqMan assay (Applied Biosystems, Foster City, CA). The TaqMan assays were read on a 7900HT real-time PCR instrument (Applied Biosystems, Foster City, CA) and genotypes were called using a semi-automated clustering routine. The common G allele encodes the glutamic acid (E) residue in the ANGPTL4 protein, while the lower frequency A allele encodes a lysine (K). Because of small numbers of A allele homozygotes, G/A heterozygotes (n = 338) and A/A homozygotes (n = 6) were combined and their values compared with those of G/G homozygotes (n = 7,939).

Statistical Analysis

Participant characteristics at the baseline exam were compared between ANGPTL4[E40 K] genotype group (SAS proc GLM). TG concentration at each of the four ARIC exams, as well as the change in TG across exams, was calculated separately by ANGPTL4[E40 K] genotype group. Changes in TGs according to 2-cm changes in waist circumference or 2-kg changes in body weight were also calculated separately by ANGPTL4[E40 K] genotype group. Interactions between genotype group and body weight change or between genotype group and waist circumference change were estimated with cross-product terms. Unadjusted and multivariable adjusted values of these outcomes were calculated. The multivariable model included study center (Minnesota, Maryland, North Carolina), age (years), sex (male/female), smoking (current/former/never smoker and cigarette years), education level (up to and including grade school, high school without diploma, high school graduate, vocational school, college graduate, graduate school/professional school), physical activity (Baecke sport score) (13), and alcohol intake (g/d).

All analyses were performed with SAS version 9.1. Because of skewed distribution, TG values were transformed to the natural log scale for analysis and were back-transformed to the original scale for presentation. When calculating the change between exam 1 and exam 4 TGs, untransformed TG values were used.


Participant demographic and lifestyle characteristics, as well as baseline body weight, BMI, and waist circumference, did not differ between ANGPTL4[E40 K] G allele homozygotes and A allele carriers with the exception of greater alcohol intake in carriers (P = 0.02) (Table 1). Plasma concentrations of TGs were 14 to 23 mg/dL lower in A allele carriers than in G allele homozygotes at each exam, but the degree of change (increase) between the baseline exam and exam 4 was similar: +17 to 19 mg/dL. Adjusting for waist circumference or BMI (change over follow-up or measured at the exam of interest) did not alter these results (data not shown). Adjustment for basic demographic and lifestyle factors (study center, age, sex, smoking status, cigarette years, physical activity, alcohol intake) did not change these associations (data not shown).

Selected baseline characteristics and plasma triglycerides across all exams in white men and women from the ARIC study stratified by ANGPTL4[E40 K] genotype*

In all participants, a 2-kg increase in body weight between baseline and exam 4 was associated with a 5.4 mg/dL increase in TGs (β ± SE: 5.4 ± 0.3), and a 2-cm increase in waist circumference was associated with a 2.3 mg/dL increase in TG (2.3 ± 0.2), adjusted for demographic and lifestyle factors. Table 2 shows the TG changes associated with 2-kg body weight and 2-cm waist circumference changes in each ANGPTL4[E40 K] genotype group. These data showed no evidence of interaction between ANGPTL4[E40 K] genotype and either measure of anthropometric change in terms of TG concentration change (p for interaction [gt-or-equal, slanted]0.30). If only baseline data were considered, there was also no evidence of interaction between ANGPTL4[E40 K] genotype and BMI, body weight, or waist circumference (p for interaction >0.71, data not shown). Adjustment for hormone use or diagnosis of diabetes over follow-up also did not change these results.

Nine-year change in plasma triglyceride concentrations associated with a 2-cm increase in waist circumference or a 2-kg increase in body weight by ANGPTL4[E40 K] genotype group

As an alternative representation, we also calculated the difference in TG change between genotype groups stratified by four weight change categories (percent change in body weight = [(exam 4 weight – exam 1 weight)/exam 1 weight]*100): 1) weight loss or no weight gain, 2) >0–5% weight gain 3) >5%–10% weight gain, and 4) >10% weight gain. Again, the change in TGs within each percentage weight change category did not significantly differ between ANGPTL4[E40 K] G allele homozygotes and A allele carriers (Table 3).

Nine-year change in plasma triglyceride concentrations by ANGPTL4[E40 K] genotype and percent weight gain category*


The current longitudinal investigation, including 8,283 white men and women from the ARIC study, showed that both body weight and ANGPTL4[E40 K] genotype independently influence plasma TG concentrations, but together genotype and body weight do not interactively determine TG concentration. Persons carrying the ANGPTL4[E40 K] A allele had lower concentrations of TGs than non-carriers at all exams, but TG concentrations increased as body weight increased uniformly in both genotype groups. These data indicate that while ANGPTL4 [E40 K] variation may confer CVD protection via maintenance of lower TG concentrations (14, 15), other environmental and lifestyle factors still operate as they would regardless of genotypic tendency. Having the “protective” genotype does not preclude the potent effects of lifestyle factors, such as increasing body weight, on a CVD risk factor like TG.

Our results are in contrast to those of previous studies investigating gene-environment (body weight) interactions with respect to lipid concentrations. Two cross-sectional studies reported interactions between measures of adiposity and the hepatic lipase 514C >T polymorphism, with both showing the T allele associated with greater high-density lipoprotein only in individuals who had a healthy body weight (BMI <25 kg/m2) (16) or less visceral adiposity (visceral adipose tissue <130 cm2) (17). Since the association between ANGPTL4[E40 K] and TGs was reported only recently, there are no published studies of the interactive effects of environmental factors and ANGPTL4[E40 K] genotype. ANGPTL4[E40 K] variation is hypothesized to decrease TG concentrations due to a loss of function of the ANGPTL4 protein, a reversal of lipoprotein lipase inhibition, and an increase in TG clearance (13). Fat mass, or changes in fat mass, reflected by various measures of adiposity also influence TG concentration via lipoprotein lipase and also through modulation of hepatic lipase activity. For example, weight loss has been shown to increase LPL mRNA (18) and BMI has been positively associated with hepatic lipase activity (10, 19), resulting in lower HDL and greater TG.

Because the mechanisms by which ANGPTL4[E40 K] variation and body weight influence TG concentrations are similar, we hypothesized that carriers of the ANGPTL4[E40 K] variant allele might be resistant to the influences of changes in adiposity (reflected here by body weight or waist circumference change). However, data showed that the positive slope of TG change associated with increasing body weight or waist circumference was similar in both genotype groups (i.e., the TG differential between genotype groups did not materially change as body weight and waist circumference changed over follow-up).

Strengths of this analysis include a large sample size and longitudinal measures of TG, body weight, and waist circumference. There are also potential limitations to this analysis. Although careful internal and external quality control procedures were followed in the current study (12), TG concentrations are measured with a certain degree of error due to intraindividual variability, which could complicate the interpretation of changes in TG concentration over time. However, this should be nondifferential. Our measure of total adiposity (body weight) also includes lean body mass and thus may not detect possible body fat differences between the genotype groups; however, use of waist circumference, which is more highly correlated with total and central body adiposity than is weight, also showed no differential effect on TG change by genotype. The methods of modeling “change” in epidemiological analyses are debated, but we believe the approach used here was most appropriate, given the strong association between genotype and TG at baseline and the potential for measurement error in the outcome of interest (TG) (20). Finally, these data were exclusively from white men and women from the United States and, therefore, cannot be generalized with confidence to other race groups or other geographical areas.

In conclusion, changes in plasma TG over time are similar between ANGPTL4[E40 K] A allele carriers and non-carriers, with A allele carriers maintaining relatively lower TG concentrations despite increases in TG concentration over 9 years of follow-up in both groups. Furthermore, ANGPTL4[E40 K] body weight and waist circumference changes impart a similar influence on TG regardless of genotype. As with all studies of gene x environment interaction, these findings require investigation in other large cohorts before definitive conclusions can be drawn, but they do suggest that to maintain low TG concentration and, thus, lower CVD risk, maintenance of healthy body weight and waist circumference are important, regardless of ANGPTL4[E40 K] genotype.

The Atherosclerosis Risk in Communities Study is carried out as a collaborative study supported by National Heart, Lung, and Blood Institute contracts N01-HC-55015, N01-HC-55016, N01-HC-55018, N01-HC-55019, N01-HC-55020, N01-HC-55021, and N01-HC-55022. The authors thank the staff and participants of the ARIC study for their important contributions.

Selected Abbreviations and Acronyms

adipokine angiopoietin-like 4 (gene)
Atherosclerosis Risk in Communities (study)


1. Li C. Genetics and regulation of angiopoietin-like proteins 3 and 4. Curr Opin Lipidol. 2006;17:152–156. [PubMed]
2. Merkel M, Eckel RH, Goldberg IJ. Lipoprotein lipase: genetics, lipid uptake, and regulation. J Lipid Res. 2002;43:1997–2006. [PubMed]
3. Yoshida K, Shimizugawa T, Ono M, Furukawa H. Angiopoietin-like protein 4 is a potent hyperlipidemia-inducing factor in mice and inhibitor of lipoprotein lipase. J Lipid Res. 2002;43:1770–1772. [PubMed]
4. Romeo S, Pennacchio LA, Fu Y, Boerwinkle E, Tybjaerg-Hansen A, Hobbs HH, Cohen JC. Population-based resequencing of ANGPTL4 uncovers variations that reduce triglycerides and increase HDL. Nat Genet. 2007;39:513–516. [PMC free article] [PubMed]
5. Cohen JC, Boerwinkle E, Mosley TH, Jr, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med. 2006;354:1264–1272. [PubMed]
6. Lloyd-Jones DM, Liu K, Colangelo LA, Yan LL, Klein L, Loria CM, et al. Consistently stable or decreased body mass index in young adulthood and longitudinal changes in metabolic syndrome components: the Coronary Artery Risk Development in Young Adults Study. Circulation. 2007;115:1004–1011. [PubMed]
7. Norman JE, Bild D, Lewis CE, Liu K, West DS. The impact of weight change on cardiovascular disease risk factors in young black and white adults: the CARDIA study. Int J Obes Relat Metab Disord. 2003;27:369–376. [PubMed]
8. Siervogel RM, Wisemandle W, Maynard LM, Guo SS, Chumlea WC, Towne B. Lifetime overweight status in relation to serial changes in body composition and risk factors for cardiovascular disease: The Fels Longitudinal Study. Obes Res. 2000;8:422–430. [PubMed]
9. Truesdale KP, Stevens J, Lewis CE, Schreiner PJ, Loria CM, Cai J. Changes in risk factors for cardiovascular disease by baseline weight status in young adults who maintain or gain weight over 15 years: the CARDIA study. Int J Obes (Lond) 2006;30:1397–1407. [PMC free article] [PubMed]
10. Despres JP, Moorjani S, Ferland M, Tremblay A, Lupien PJ, Nadeau A, et al. Adipose tissue distribution and plasma lipoprotein levels in obese women. Importance of intra-abdominal fat. Arteriosclerosis. 1989;9:203–210. [PubMed]
11. Nagele U, Hagele EO, Sauer G, Widemann E, Lehmann P, Wahlefeld AW, Gruber W. Reagent for the enzymatic determination of serum total triglycerides with improved lipolytic efficiency. J Clin Chem Clin Biochem. 1984;22:165–174. [PubMed]
12. Sharrett AR, Chambless LE, Heiss G, Paton CC, Patsch W. Association of postprandial triglyceride and retinyl palmitate responses with asymptomatic carotid artery atherosclerosis in middle-aged men and women. The Atherosclerosis Risk in Communities (ARIC) Study. Arterioscler Thromb Vasc Biol. 1995;15:2122–2129. [PubMed]
13. Baecke JA, Burema J, Frijters JE. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr. 1982;36:936–942. [PubMed]
14. Ginsberg HN. New perspectives on atherogenesis: role of abnormal triglyceride-rich lipoprotein metabolism. Circulation. 2002;106:2137–2142. [PubMed]
15. McBride PE. Triglycerides and risk for coronary heart disease. JAMA. 2007;298:336–338. [PubMed]
16. Zhang C, Lopez-Ridaura R, Rimm EB, Rifai N, Hunter DJ, Hu FB. Interactions between the -514C->T polymorphism of the hepatic lipase gene and lifestyle factors in relation to HDL concentrations among US diabetic men. Am J Clin Nutr. 2005;81:1429–1435. [PubMed]
17. 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 >T polymorphism on plasma HDL-cholesterol levels in French-Canadian men. Mol Genet Metab. 2003;78:31–36. [PubMed]
18. Patalay M, Lofgren IE, Freake HC, Koo SI, Fernandez ML. The lowering of plasma lipids following a weight reduction program is related to increased expression of the LDL receptor and lipoprotein lipase. J Nutr. 2005;135:735–739. [PubMed]
19. Nie L, Wang J, Clark LT, Tang A, Vega GL, Grundy SM, Cohen JC. Body mass index and hepatic lipase gene (LIPC) polymorphism jointly influence postheparin plasma hepatic lipase activity. J Lipid Res. 1998;39:1127–1130. [PubMed]
20. Glymour MM, Weuve J, Berkman LF, Kawachi I, Robins JM. When is baseline adjustment useful in analyses of change? An example with education and cognitive change. Am J Epidemiol. 2005;162:267–278. [PubMed]