- Email: [email protected]
The effect of the APOE polymorphism on HDL-C concentrations depends on the cholesterol ester transfer protein gene variation in a Southern European population
Clinica Chimica Acta 366 (2006) 196 – 203 www.elsevier.com/locate/clinchim
The effect of the APOE polymorphism on HDL-C concentrations depends on the cholesterol ester transfer protein gene variation in a Southern European population Jose´ V. Sorlı´ *, Dolores Corella, Francesc France´s, Judith B. Ramı´rez, Jose´ I. Gonza´lez, Marisa Guille´n, Olga Portole´s Department of Preventive Medicine and Public Health, Genetic and Molecular Epidemiology Unit, School of Medicine, University of Valencia, Avda. Blasco Iban˜ez, 15, 46010 Valencia, Spain Received 5 October 2005; received in revised form 5 October 2005; accepted 5 October 2005 Available online 19 January 2006
Abstract Background: Apolipoprotein E (ApoE) locus has consistently shown a significant association with low-density lipoprotein cholesterol (LDLC). However, its impact on high-density lipoprotein cholesterol (HDL-C) has been highly controversial suggesting that it may be contextdependent. We examined the gene – gene interaction between the common ApoE and the CETP polymorphisms in determining HDL-C concentrations in men and women from the general population. Methods: 550 unrelated Caucasian subjects were randomly selected from a Mediterranean Region in Spain. Plasma lipids, anthropometric, clinical and lifestyle variables were measured. Common ApoE and CETP-TaqIB polymorphisms were determined. Results: We have found a gene – gene interaction between and ApoE and the CETP loci in determining HDL-C concentrations. Thus, after adjustment for gender, age, body mass index, tobacco smoking, alcohol consumption, physical exercise and medication, carriers of the E4 allele had lower HDL-C concentrations [mean and (standard error): 40.1 (2.6) mg/dL] than E2 subjects [47.7 (3.2) mg/dL; p = 0.019], and even lower than those of the E3 subjects [44.7 (1.4) mg/dL; p = 0.042], only if they had the B1B1 genotype. However, mean HDL-C concentrations were higher among those with E4 allele carrying the B2 allele at the CETP gene locus [50.5 (2.3) mg/dL], and lower among E2 subjects carrying the B2 allele [45.5 (2.6) mg/dL]. This interaction was observed in both men and women. This gene – gene interaction remained statistically significant even after additional adjustment for triglycerides. Conclusions: The effect of the ApoE polymorphism on HDL-C concentrations depends on the CETP polymorphism, explaining some of the controversial results previously reported for this polymorphism. D 2005 Elsevier B.V. All rights reserved. Keywords: ApoE; CETP; Polymorphism; Gene – gene interaction; HDL-C
1. Introduction Understanding the relationship between relevant DNA polymorphisms and plasma lipid concentrations is expected to improve the diagnosis, treatment and prevention of cardiovascular diseases (CVD) [1]. Until recently, most of the association studies have focused on elucidat-
* Corresponding author. Tel.: +34 96 3864417; fax: +34 96 3864166. E-mail address: [email protected] (J.V. Sorlı´). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2005.10.001
ing the role of only one or a few polymorphisms in one gene at a time. There is increasing evidence that indicates the genetic polymorphisms contributing to CVD risk do not have the same effects in different environments, or in different genetic backgrounds [2– 5]. In general, these context-dependent genetic effects may involve the socalled gene – environment interactions as well as the gene– gene interactions or epistasis [6,7]. Most previous studies of the genetics of plasma lipid concentrations fail to take context-dependent effects into account. In this regard, there is a growing awareness that the failure to
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
replicate single-locus association studies for lipid metabolism may be due to the fact that gene –environment and gene – gene interactions are the norm rather than the exception [3,7]. Moreover, the few studies analyzing context-dependent effects on lipid traits have focused on environmental factors and little is known about gene – gene interactions [8 –10]. The apolipoprotein E (ApoE) gene (ApoE) is one of the most well studied genes to date. ApoE mediates the catabolism of chylomicron and very low density lipoprotein (VLDL) remnants via a remnant receptor, and the binding of chylomicron remnants, VLDL, and intermediate density lipoproteins to the low density lipoprotein (LDL) receptor [11]. A common three-allele (E2/E3/E4) polymorphism in the ApoE gene, identified in more than 100 different populations worldwide, is potentially one of the most important genetic predictors of plasma LDL-cholesterol (LDL-C) concentrations [11 –15]. The well-documented pattern of increasing LDL-C concentrations from E2 to E3 to E4 across different populations, has been called the invariant effect of this polymorphism [16]. In contrast, the associations between the ApoE polymorphism and plasma levels of high-density lipoprotein (HDL) cholesterol (HDL-C), have been controversial [15,17 – 19], suggesting strong context dependency. Recently, Djousse et al. [20] have reported a gene – environment interaction between the ApoE polymorphism and alcohol intake on plasma HDL-C concentrations in men and women who participated in the National Heart, Lung, and Blood Institute Family Heart Study. However, gene – gene interactions were not examined in this study. In examining gene –gene interactions in determining HDL-C concentrations, the crucial challenge is to select another relevant gene that is functionally linked to the former in the corresponding metabolic pathway and, for which, biological plausibility of functional interactions exit. In this regard, the enzyme cholesteryl ester transfer protein (CETP) plays a key role in the reverse cholesterol transport; the transport of cholesterol from cells of the arterial wall to the liver, which is mediated by ApoE-containing HDL particles [21,22]. ApoE enhances the lipid exchange between lipoproteins mediated by CETP, and alterations in CETP activity contributes to changes in HDL composition and size [23,24]. As both ApoE and CETP contribute to the reverse cholesterol transport, the effects of variations at the CETP locus may very well interact with the ApoE genotype. Several polymorphisms have been described in the CETP gene [25]. Among them, the CETP-TaqIB polymorphism is the variant that has been most firmly associated with HDL-C concentrations in several populations [26]. Therefore, our aim was to investigate the possible gene –gene interaction between the ApoE variation and the CETPTaqIB polymorphism in determining HDL-C concentrations in a randomly selected Southern European population from Spain.
197
2. Materials and methods 2.1. Subjects and study design In the present study, 550 unrelated Caucasian subjects (171 men and 379 women) from the general population were enrolled. All subjects (aged 18 –85 years) were randomly recruited from the Valencia Region on the East Mediterranean coast of Spain. The study protocol was approved by the Ethics Committee on Human Research of Valencia University. We present data obtained from the individuals attending the first cross-sectional examination carried out between February 2000 and January 2002, who were selected as previously reported [27]. The random selection and recruitment of the study subjects was carried out at the Primary Healthcare Center in Paiporta (Valencia). All the study participants were volunteers and gave their informed consent. Previously validated questionnaires were distributed at the time of the medical examination, and participants were invited to fill them in. A venous blood sample was collected during the medical check-up [27]. Alcohol consumption was carefully evaluated and drinkers (subjects with any amount of alcohol consumed) and nondrinkers (alcohol consumption = 0) were considered. Physical activity was estimated from questions about regular leisure-time physical sports and subjects were categorized as sedentary (no regular physical exercise), or active (regular physical exercise). 2.2. Anthropometric, clinical and biochemical data Anthropometric measurements were taken using standard techniques: weight with light clothing by digital scales; height without shoes by fixed stadiometer. Body mass index (BMI) was calculated as weight (kg)/height (m2). Blood pressure was taken with a calibrated mercury sphygmomanometer following the WHO-MONICA (World Health Organization-Multi-national monitoring of trends and determinants in cardiovascular disease) protocol [27]. Participants were instructed to fast for at least 12 h before a morning examination. Plasma total cholesterol and triglycerides were determined by a Technicon Chem 1 assay, and HDL-C was measured in the supernatant after precipitation of apolipoprotein B-containing lipoproteins with heparin-manganese chloride as previously described [27]. LDL-C was calculated according to the Friedewald et al. equation [28] for samples with triglyceride concentrations below 400 mg/dL. 2.3. DNA extraction and genotyping Genomic DNA was isolated from white blood cells by phenol –cloroform extraction and ethanol precipitation [27]. DNA samples were subjected to amplification by the polymerase chain reaction (PCR) in an Eppendorf DNA thermal cycler. For the common ApoE polymorphism,
198
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
Table 1 Demographic, biochemical and lifestyle characteristics of the study subjects Men (n = 171) Women (n = 378) p*
Age (years) Weight (kg) Height (m) BMI (kg/m2) Waist (cm) Hip (cm) Waist-to-hip ratio Glucose (mg/dL) Uric acid (mg/dL) Total-C (mg/dL) LDL-C (mg/dL) HDL-C (mg/dL) Triglycerides (mg/dL) SBP (mm Hg) DBP (mm Hg) Current smokers (%) Alcohol users (%) Taking lipid lowering drugs (%) Regular physical exercise (%)
Mean (S.D.)
Mean (S.D.)
45.1 (18.8) 79.0 (13.5) 1.70 (0.08) 27.2 (4.1) 95.6 (11.7) 99.8 (8.3) 0.96 (0.08) 110.7 (35.2) 5.1 (1.3) 206.9 (42.9) 136.3 (35.0) 44.2 (8.8) 154.6 (94.6) 131.6 (19.9) 79.3 (10.8) 38.1 79.6 7.0 18.7
45.9 (17.2) 68.8 (14.2) 1.57 (0.07) 28.2 (5.8) 88.6 (15.2) 101.4 (12.8) 0.87 (0.07) 103.2 (24.6) 3.6 (1.1) 208.3 (41.4) 136.3 (36.9) 52.4 (13.9) 118.8 (72.3) 127.2 (23.3) 77.4 (12.3) 24.6 54.8 9.3 13.3
0.648 <0.001 <0.001 0.113 <0.001 0.219 <0.001 0.005 <0.001 0.739 0.992 <0.001 <0.001 0.081 0.147 0.007 <0.001 0.379 0.121
S.D.: standard deviation, BMI: body mass index, Total-C: total cholesterol, LDL-C: low density lipoprotein cholesterol, HDL-C: high density lipoprotein cholesterol, SBP: systolic blood pressure, DBP: diastolic blood pressure. * p-value obtained in the comparison between men and women: Student’s t-test or chi-squared test.
genotyping was carried out by the method of Hixson and Vernier [29]. The CETP-TaqIB polymorphism was determined as previously indicated [24]. 2.4. Statistical analysis Normal distribution for all continuous variables was checked. Triglycerides, were skewed and this variable was logarithmically transformed to improve normality. To assess mean differences between genders, Student’s t-test was used. Allele frequencies were estimated by gene counting and v 2 tests were used to test differences between observed and expected frequencies, assuming Hardy – Weinberg equilibrium, and to test differences in percentages between men and women. For multiple comparisons of means between genotypes, one-way analysis of variance was performed. Bonferroni test was applied for multiple comparisons. Men and women were analyzed together after checking the homogeneity of the genotype effect. To control for potential confounders, mean values were adjusted for age, gender, BMI, tobacco smoking, alcohol consumption, physical exercise and medication by multivariate models. Multivariate linear regression analysis with dummy variables for categorical and interaction terms was used to test the null hypothesis of no interaction between the CETP and the ApoE polymorphism after adjustment for potential confounders. Homogeneity of allelic effects according to gender was tested by introducing the corresponding second-order interaction term in the regression model. Standard regres-
sion diagnostic procedures were used to ensure the appropriateness of these models. Analyses were done using the Statistical Package of Social Sciences (SPSS, version 11.0) for Windows.
3. Results Participants were genotyped for the ApoE polymorphism. One woman with the E2/E4 genotype was excluded for the subsequent analyses. Demographic, biochemical, and lifestyle characteristics of the 549 study subjects (171 men and 378 women) are presented in Table 1. The mean age was 45.6 years, with no statistical differences between men and women. All of them were also genotyped for the TaqIB polymorphism at the CETP gene locus. Table 2 shows genotypes for the CETP variants according to the ApoE polymorphism. No differences by gender were detected, and men and women were analyzed together. Genotype frequencies did not deviate from the Hardy –Weinberg equilibrium. Subjects with the E2/E2 and E2/E3 genotypes were grouped as E2, subjects E3/E3 were classified as E3, and E3/E4 or E4/ E4 were grouped as E4. Table 3 shows crude (unadjusted) and adjusted means of HDL-C and triglycerides according to the ApoE and the CETP polymorphisms in the study subjects. Individuals with the B1B2 genotype were grouped with those of the B2B2 genotype as B2 carriers. No statistically significant differences of HDL-C concentrations per ApoE groups were found ( P crude = 0.881; P adjusted = 0.851). Triglyceride concentrations were significantly different by the ApoE polymorphism in the crude model, but differences were borderline significant ( p = 0.058) when the adjustment by age, gender, BMI, tobacco smoking, alcohol consumption, physical exercise and lipid lowering medication was accounted for. In addition, taking into account that the difference in plasma triglycerides between the ApoE genotypes may mask any effect of the ApoE polymorphism on HDL-C concentrations, an adjustment for triglycerides was carried out . However, this adjustment did not affect the previous results (Table 3). On the other hand, the CETP polymorphism was significantly associated with HDL-C. Carriers of the B2 allele had higher HDL-C concentrations ( p = 0.034) than B1B1 homozygotes, and this difference remained statistically significant ( p = 0.002) even after adjustment for covariates. No association of the CETP polymorphism with triglyceride concentrations was found. Additionally, when the association between the ApoE and the Table 2 APOE genotype distribution according to the CETP polymorphism CETP
B1/B1 B2 carrier
APOE genotype E2, n (%)
E3, n (%)
E4, n (%)
22 (10.4) 29 (9.4)
160 (75.5) 264 (76.4)
30 (14.1) 44 (14.2)
The APOE genotype distribution was not significantly different among the CETP genotypes ( p = 0.707).
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
199
Table 3 Crude and adjusted means of HDL-C and triglyceride concentrations depending on the APOE and the CETP polymorphism Genotype
HDL-C (mg/dL)
Triglycerides (mg/dL)
Crude (S.D.)
Adjusted* (S.E.)
Crude (S.D.)
Adjusted* (S.E.)
APOE E2 (n = 51) E3 (n = 424) E4 (n = 74) P
49.7 (10.3) 49.8 (13.4) 48.8 (13.1) 0.881
46.2 (2.2) 46.9 (1.2) 46.0 (1.8) 0.851**
157.9 (96.6) 124.1 (73.8) 142.4 (104.3) 0.042
176.3 (14.8) 140.4 (7.8) 155.0 (11.1) 0.058
CETP B1/B1 (n = 212) B2 carrier (n = 337) P
47.9 (10.9) 50.9 (14.2) 0.034
44.3 (1.4) 48.5 (1.2) 0.002
134.9 (92.2) 126.3 (73.6) 0.553
155.7 (8.9) 139.3 (8.0) 0.175
S.D.: standard deviation, S.E.: standard error. P: P-value obtained in the global comparison between genotypes (ANOVA and ANCOVA tests). * Mean value adjusted for: gender, age, BMI, medical treatment, tobacco smoking, alcohol consumption, and physical activity. ** After additional adjustment for triglycerides, the adjusted P value was non-significant ( P = 0.840) and the adjusted mean values were: 48.4 (1.9); 47.5 (0.8) and 48.2 (1.6) for E2, E3 and E4 genotypes, respectively.
CETP polymorphisms with HDL-C and triglycerides was examined, after excluding those subjects taking lipid lowering drugs (n = 12 men and 35 women), no differences in the statistical significance of the results were found. In addition, no heterogeneity of the effect of the ApoE polymorphism or the CETP polymorphism on these lipid traits by gender was found ( p > 0.05 for all the four gene – gender interactions). Furthermore, the gene –gene interaction between and ApoE and the CETP loci in determining HDL-C and triglyceride concentrations was investigated. For HDL-C concentrations, a statistically significant interaction term was found ( p = 0.047) after adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and lipid lowering medication. The effect of the ApoE polymorphism on HDL-C concentrations depends on the CETP polymorphism. Thus, carriers of the E4 allele had lower HDL-C concentrations [mean and (standard error): 40.1 (2.6) mg/dL] than E2 subjects [47.7 (3.2) mg/dL; p = 0.019], and even lower than those of the E3 subjects [44.7 (1.4) mg/dL; p = 0.042], only if they had the B1B1 genotype. However, mean HDL-C concentrations were higher among those with E4 allele carrying the B2 allele at the CETP gene locus [50.5 (2.3) mg/dL], and lower among E2 subjects carrying the B2 allele [45.5 (2.6) mg/dL]. No heterogeneity of the ApoE –CETP interaction by gender was found. The statistical significance of the interaction term between ApoE – CETP – gender in the multivariate hierarchical regression model was p = 0.296. Moreover, when the homogeneity of the gene –gene interaction was explored in men and women separately, the direction of the interaction effect was the same in both genders. Because of the small number of men in comparison with women, the ApoE – CETP term was statistically significant in women ( p = 0.035), and borderline significant in men. In addition, as 45% of women were post-menopausal, we tested the heterogeneity of the gene – gene interaction effect by
menopausal status and similar results were found for preand post-menopausal women. We further studied the interaction between the ApoE and CETP polymorphisms in determining HDL-C concentrations after the exclusion of subjects taking lipid lowering drugs. In the multivariate model, neither the statistical significance of the interaction term between the ApoE and the CETP polymorphism ( p = 0.047), nor the magnitude of the effect were modified in this sub-sample, confirming the consistency of the interaction. Moreover, to discard the influence of triglyceride concentrations on this interaction, an additional adjustment for plasma triglycerides was carried out. Fig. 1 shows adjusted means for plasma HDL-C depending on the ApoE and the CETP polymorphisms after adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and triglycerides in subjects who were not taking lipid lowering drugs. Finally, the ApoE – CETP interaction in determining triglyceride concentrations was assessed (results not shown). No statistically significant interaction between the ApoE and the CETP polymorphism was observed ( p = 0.590). However, a decrease in the triglyceride concentrations was observed in E4 subjects carrying the B2 allele as compared with the other groups (140.1 T13.6 mg/dL in E4 subjects carrying the B2 allele versus 179.1 T16.1 mg/dL in E4 subjects with the B1B1 genotype; p = 0.247). No heterogeneity of the ApoE – CETP interaction by gender was found in determining triglyceride concentrations ( p = 0.828 for the term ApoE –CETP – gender in the multivariate hierarchical regression model). Although the decrease in triglyceride concentrations in E4 subjects depending on the CETP polymorphism did not reach statistical significance, the concomitant increase of the HDL-C concentrations in carriers of the B2 allele made the triglycerides to HDL-C ratio reach statistical significance in this group. Thus, E4 subjects carrying the B2 allele had a triglycerides to HDL-C ratio statistically lower than E4 homozygous for the B1
200
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
56
P for interaction
54
b
APOE*CETP=0.04 APOE*CETP=0.049
HDL-C (mg/dL) H
52 50
a
48 46
A APOE genotype 44
E2 a b
42
E3 E4
40 B1/B1
B2 carrier
CETP polymorphism Fig. 1. Adjusted means for plasma HDL-C depending on the ApoE and the CETP polymorphisms in men (n = 159) and women (n = 343) who were not taking lipid lowering medications. Mean values were adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and triglycerides. P value for the interaction term was estimated from the multivariate regression model. P values for the comparison between two groups were obtained after multivariate adjustment and correction for multiple comparisons (Bonferroni test). Values marked with the same letter are statistically ( P < 0.05) different. Error bars: standard error of means.
allele: 3.1 (95% CI: 2.1 –4.1) versus 5.3 (95% CI: 4.2– 6.4), respectively.
4. Discussion In this study, we have found a statistically significant gene – gene interaction between the common ApoE polymorphism and the CETP-TaqIB polymorphism in determining plasma HDL-C concentrations in a random sample of Caucasian adults from the general population. This finding may help to explain the controversial results previously reported by dozens of researchers concerning the association between the ApoE polymorphism and HDLC. In contrast to the well-documented association between the common ApoE polymorphism and LDL-C concentrations [11], the association between this polymorphism and HDL-C concentrations has been highly inconsistent [12 – 19]. Thus, although in the majority part of studies no statistically significant association between the common ApoE polymorphism and HDL-C concentrations was observed [12,14,17,30], there were some investigations reporting that HDL-C concentrations tended to be statistically lower in E4 and higher in E2 carriers [15,18,31]. In other studies, only the significant association between the E4 allele and lower HDL-C concentrations was found [32,33]. Moreover, in a review of the ApoE protein, Mahley and Rall [34] reported that, although the effect of the ApoE variation on HDL-C was not significant, a pattern of
increasing HDL-C concentrations from E2 to E3 to E4 alleles can be observed by combining several studies. In the present study, we did not find any significant association between the ApoE polymorphism and HDL-C concentrations when the effect of this polymorphism was analyzed alone. However, when the gene –gene interaction with the CETP-TaqIB polymorphism was considered, we found statistically significant associations between the ApoE polymorphism and HDL-C depending on the CETP genotypes. Thus, in B1B1 subjects, carriers of the E4 allele had the lowest HDL-C concentrations, whereas carriers of the E2 allele had the highest. However, in subjects carrying the B2 allele, this trend was reversed, and mean HDL-C concentrations were higher among those with the E4 allele. In addition, no heterogeneity of the ApoE –CETP interaction by gender was found, decreasing the likelihood that this result has been obtained by chance. These results are in agreement with Miltiadous et al. [35] who studied 200 normolipidemic individuals from Greece and found that participants carrying the E4 allele, the B1 allele of the CETP-TaqIB polymorphism and the allele T of the ApoA4 gene (A to T polymorphism at site 347) had statistically significant lower HDL-C concentrations compared to those not carrying the above allele combination). Although the physiological mechanism that may explain the observed results needs to be elucidated, there are some interesting features related to these polymorphism that may contribute to this interaction. The TaqIB polymorphism has been associated with CETP activity. Thus, in the Framingham
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
Study male and female carriers of the B2 allele had significantly lower CETP activity than did those homozygotes for the B1 allele, supporting the observation that lower CETP activity is associated with increased HDL-C concentrations [24]. In the present study, before the interaction with the ApoE gene was analyzed, we also found that the B2 allele was associated with higher HDL-C concentrations in line with the majority of studies [26]. The mechanism by which TaqIB polymorphism may affect CETP activity is not known. It is unlikely that this polymorphism located in an intron represents a functional mutation. However this polymorphism may have an effect on the transcriptional regulation of the CETP gene or it could be a marker of a still unknown functional mutation [36]. On the other hand, preliminary experimental studies have shown that ApoE increases CETP activity suggesting that the effects of ApoE on the changes of lipid profile in the HDL fraction are not the direct effect of ApoE on lipid transfer, but mediated by CETP activity in the plasma [37]. Several studies have reported that plasma ApoE concentrations are determined in part by the ApoE polymorphism, thus the E2 allele is associated with higher ApoE concentrations; conversely, the E4 allele is related to lower ApoE concentrations [38,39]. These observations are consistent with the faster rate of catabolism of particles containing ApoE4 compared with those containing ApoE3
201
[40]. In some studies examining the effect of the ApoE genotype on CETP activity, a significant interaction between the change in plasma CETP and the ApoE genotype in response to cholesterol feeding has been reported [41]. Thus, Martin et al. [41] observed that E2 subjects responded to cholesterol feeding with a increase of plasma CETP and lower plasma HDL-C, whereas E4 subjects showed a lower increase of plasma CETP and higher plasma HDL-C concentrations. E3 subjects had an intermediate effect between E2 and E4. This situation is similar to the effect that we observed by ApoE genotype in the case of B2 allele. Moreover, taking into account that it has been demonstrated that HDL particle size in carriers of the B2 allele is higher than in B1B1 subjects [24], and that ApoE is essential for the formation of large HDL particles [42,43], it seems reasonable to hypothesise that carriers of the B2 allele have HDL particles more rich in ApoE than B1B1 homozygotes. Considering that the acquisition of ApoE by HDL enables the lipoproteins to interact with LDL and ApoE receptors [44], we suggest that the number of ApoE in HDL is crucial for the receptor interaction. In B2 subjects, the buoyant and enriched-ApoE HDL allows the efficient interaction with the receptors (Fig. 2), independently of the ApoE-receptor affinity determined by the ApoE genotype [19]. Therefore, in carriers of the B2 allele, HDL is efficiently presented and
Fig. 2. Proposed model of the interaction between ApoE and CETP depending on their specific genetic variants. Panel A shows the situation predicted in individuals carrying the B2 allele at the CETP locus. In these subjects, the buoyant and ApoE-enriched HDL particles allow the efficient interaction between ApoE and the scavenger receptor class B-type I (SR-BI), the stimulatory effect of the different ApoE isoforms on the CETP activity prevailing over the different affinity of the ApoE isoforms for the SR-BI as depicted in the figure. Thus, higher plasma HDL-C concentration can be observed in carriers of the E4 allele compared to the E2. Conversely, for CETP B1 allele homozygous subjects (panel B), the small and impoverished HDL have a limited capability of being presented to SR-BI. In this situation, therefore, the differing affinity of the different ApoE isoforms for the receptor prevails, which is higher in E4 allele carriers than those of E2 or E3.
202
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
attracted by the HDL receptor, scavenger receptor BI [45,46], the stimulatory effect of ApoE on CETP activity prevailing over the ApoE-receptor affinity. On the other hand, in B1B1 subjects the smaller and poor-ApoE HDL particles may limit the capacity to present HDL to the scavenger receptor BI. For this reason, in B1B1 subjects, efficient presentation may depend on the differing affinities between the type of ApoE and the receptor, that was previously concealed because the sufficient amount of apolipoprotein in the B2 situation. In conclusion, we have observed a statistically significant interaction between the ApoE and the CETP-TaqIB polymorphisms in determining plasma HDL-C concentrations in men and women from the general population. This interaction is compatible with the previously reported results obtained in experimental models and may explain some of the inconsistencies regarding the association between the ApoE polymorphism and plasma HDL-C concentrations and even cardiovascular risk. Therefore, this gene – gene interaction should be examined in other population studies to confirm consistency and to explore the role of additional genetic or environmental modulating factors. Acknowledgments This study was supported by grants (FIS PI02/1096) and (G03/160) from the Instituto de Salud Carlos III and grant GR04/043 from the Oficina de Ciencia y Tecnologı´a, Generalitat Valenciana. References [1] Ordovas JM. Cardiovascular disease genetics: a long and winding road. Curr Opin Lipidol 2003;14:47 – 54. [2] Talmud PJ, Humphries SE. Gene: environment interactions and coronary heart disease risk. World Rev Nutr Diet 2004;93: 29 – 40. [3] Hunter DJ. Gene – environment interactions in human diseases. Nat Rev Genet 2005;6(4):287 – 98. [4] Ordovas JM, Corella D, Demissie S, et al. Dietary fat intake determines the effect of a common polymorphism in the hepatic lipase gene promoter on high-density lipoprotein metabolism: evidence of a strong dose effect in this gene – nutrient interaction in the Framingham Study. Circulation 2002;106:2315 – 21. [5] Rontu R, Karhunen PJ, Ilveskoski E, et al. Smoking-dependent association between paraoxonase 1 M/L55 genotype and coronary atherosclerosis in males: an autopsy study. Atherosclerosis 2003;171: 31 – 7. [6] Klos KL, Kardia SL, Hixson JE, et al. Linkage analysis of plasma ApoE in three ethnic groups: multiple genes with context-dependent effects. Ann Hum Genet 2005;69:157 – 67. [7] Xin X, Srinivasan SR, Chen W, Boerwinkle E, Berenson GS. Interaction effect of Serine447Stop variant of the lipoprotein lipase gene and C-514T variant of the hepatic lipase gene on serum triglyceride levels in young adults: the Bogalusa Heart Study. Metabolism 2003;52:1337 – 42. [8] Tiret L. Gene – environment interaction: a central concept in multifactorial diseases. Proc Nutr Soc 2002;61:457 – 63. [9] Ordovas JM, Corella D. Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71 – 118.
[10] Moore JH, Williams SM. New strategies for identifying gene – gene interactions in hypertension. Ann Med 2002;34:88 – 95. [11] Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 1988;8:1 – 21. [12] Schaefer EJ, Lamon-Fava S, Johnson S, et al. Effects of gender and menopausal status on the association of apolipoprotein E phenotype with plasma lipoprotein levels. Results from the Framingham Offspring Study. Arterioscler Thromb 1994;14:1105 – 13. [13] Boer JM, Ehnholm C, Menzel HJ, et al. Interactions between lifestylerelated factors and the ApoE polymorphism on plasma lipids and apolipoproteins. The EARS Study. European Atherosclerosis Research Study. Arterioscler Thromb Vasc Biol 1997;17:1675 – 81. [14] Hsueh WC, Mitchell BD, Hixson JE, Rainwater DL. Effects of the ApoE polymorphism on plasma lipoproteins in Mexican Americans. Ann Epidemiol 2000;10:524 – 31. [15] Tan CE, Tai ES, Tan CS, et al. APOE polymorphism and lipid profile in three ethnic groups in the Singapore population. Atherosclerosis 2003;170:253 – 60. [16] Frikke-Schmidt R. Context-dependent and invariant associations between APOE genotype and levels of lipoproteins and risk of ischemic heart disease: a review. Scand J Clin Lab Invest Suppl 2000;233:3 – 25. [17] Pablos-Mendez A, Mayeux R, Ngai C, Shea S, Berglund L. Association of ApoE polymorphism with plasma lipid levels in a multiethnic elderly population. Arterioscler Thromb Vasc Biol 1997;17:3534 – 41. [18] Kataoka S, Robbins DC, Cowan LD, et al. Apolipoprotein E polymorphism in American Indians and its relation to plasma lipoproteins and diabetes. The Strong Heart Study. Arterioscler Thromb Vasc Biol 1996;16:918 – 25. [19] Frikke-Schmidt R, Nordestgaard BG, Agerholm-Larsen B, Schnohr P, Tybjaerg-Hansen A. Context-dependent and invariant associations between lipids, lipoproteins, and apolipoproteins and apolipoprotein E genotype. J Lipid Res 2000;41:1812 – 22. [20] Djousse L, Pankow JS, Arnett DK, Eckfeldt JH, Myers RH, Ellison RC. Apolipoprotein E polymorphism modifies the alcohol – HDL association observed in the National Heart, Lung, and Blood Institute Family Heart Study. Am J Clin Nutr 2004;80:1639 – 44. [21] de Grooth GJ, Klerkx AH, Stroes ES, Stalenhoef AF, Kastelein JJ, Kuivenhoven JA. A review of CETP and its relation to atherosclerosis. J Lipid Res 2004;45:1967 – 74. [22] Kinoshita M, Arai H, Fukasawa M, et al. Apolipoprotein E enhances lipid exchange between lipoproteins mediated by cholesteryl ester transfer protein. J Lipid Res 1993;34:261 – 8. [23] Korhonen T, Hannuksela ML, Seppanen S, Kervinen K, Kesaniemi MJ, Savolainen MJ. The effect of the apolipoprotein E phenotype on cholesteryl ester transfer protein activity, plasma lipids and apolipoprotein A I levels in hypercholesterolaemic patients on colestipol and lovastatin treatment. Eur J Clin Pharmacol 1999;54:903 – 10. [24] Ordovas JM, Cupples LA, Corella D, et al. Association of cholesteryl ester transfer protein-TaqIB polymorphism with variations in lipoprotein subclasses and coronary heart disease risk: the Framingham study. Arterioscler Thromb Vasc Biol 2000;20:1323 – 9. [25] Boekholdt SM, Thompson JF. Natural genetic variation as a tool in understanding the role of CETP in lipid levels and disease. J Lipid Res 2003;44:1080 – 93. [26] Boekholdt SM, Sacks FM, Jukema JW, et al. Cholesteryl ester transfer protein TaqIB variant, high-density lipoprotein cholesterol levels, cardiovascular risk, and efficacy of pravastatin treatment: individual patient meta-analysis of 13,677 subjects. Circulation 2005;111:278 – 87. [27] Sorli JV, Velert R, Guillen M, et al. Effects of the apolipoprotein E polymorphism on plasma lipid levels and cardiovascular disease risk in a Mediterranean population. Med Clin (Barc) 2002;118: 569 – 74. [28] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma,
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
without use of the preparative ultracentrifuge. Clin Chem 1972;18: 499 – 502. Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990;31: 545 – 8. Muros M, Rodriguez-Ferrer C. Apolipoprotein E polymorphism influence on lipids, apolipoproteins and Lp(a) in a Spanish population underexpressing ApoE4. Atherosclerosis 1996;121:13 – 21. Tiret L, de Knijff P, Menzel HJ, Ehnholm C, Nicaud V, Havekes LM. ApoE polymorphism and predisposition to coronary heart disease in youths of different European populations. The EARS Study, European Atherosclerosis Research Study. Arterioscler Thromb 1994;14:1617 – 24. Gomez-Coronado D, Alvarez JJ, Entrala A, Olmos JM, Herrera E, Lasuncion MA. Apolipoprotein E polymorphism in men and women from a Spanish population: allele frequencies and influence on plasma lipids and apolipoproteins. Atherosclerosis 1999;147:167 – 76. Dallongeville J, Lussier-Cacan S, Davignon J. Modulation of plasma triglyceride levels by ApoE phenotype: a meta-analysis. J Lipid Res 1992;33:447 – 54. Mahley RW, Rall Jr SC. Apolipoprotein E: far more than a lipid transport protein. Annu Rev Genomics Hum Genet 2000;1: 507 – 37. Miltiadous G, Hatzivassiliou M, Liberopoulos E, et al. Gene polymorphisms affecting HDL-cholesterol levels in the normolipidemic population. Nutr Metab Cardiovasc Dis 2005;15:219 – 24. Thompson JF, Durham LK, Lira ME, Shear C, Milos PM. CETP polymorphisms associated with HDL cholesterol may differ from those associated with cardiovascular disease. Atherosclerosis 2005; 181:45 – 53. Kinoshita M, Arai H, Fukasawa M, et al. Apolipoprotein E enhances lipid exchange between lipoproteins mediated by cholesteryl ester transfer protein. J Lipid Res 1993;34:261 – 8.
203
[38] Schiele F, De Bacquer D, Vincent-Viry M, et al. Apolipoprotein E serum concentration and polymorphism in six European countries: the ApoEurope Project. Atherosclerosis 2000;152:475 – 88. [39] Vincent-Viry M, Schiele F, Gueguen R, Bohnet K, Visvikis S, Siest G. Biological variations and genetic reference values for apolipoprotein E serum concentrations: results from the STANISLAS cohort study. Clin Chem 1998;44:957 – 65. [40] Davignon J, Cohn JS, Mabile L, Bernier L. Apolipoprotein E and atherosclerosis: insights from animal models. Clin Chim Acta 1999;286:115 – 43. [41] Martin LJ, Connelly PW, Nancoo SD, et al. Cholesterol ester transfer protein and high density lipoprotein responses to cholesterol feeding in men: relationship to apolipoprotein E genotype. J Lipid Res 1993; 34:437 – 46. [42] Krimbou L, Marcil M, Chiba H, Genest Jr J. Structural and functional properties of human plasma high density-sized lipoprotein containing only ApoE particles. J Lipid Res 2003;44:884 – 92. [43] Jiang XC, Beyer TP, Li Z, et al. Enlargement of high density lipoprotein in mice via liver receptor activation requires apolipoprotein E and is abolished by cholesteryl ester transfer protein expression. J Biol Chem 2003;278:49072 – 8. [44] Dergunov AD, Smirnova EA, Merched A, et al. Structural peculiarities of the binding of very low density lipoproteins and low density lipoproteins to the LDL receptor in hypertriglyceridemia: role of apolipoprotein E. Biochim Biophys Acta 2000;1484:29 – 40. [45] Arai T, Rinninger F, Varban L, et al. Decreased selective uptake of high density lipoprotein cholesteryl esters in apolipoprotein E knockout mice. Proc Natl Acad Sci U S A 1999;96:12050 – 5. [46] Varban ML, Rinninger F, Wang N, et al. Targeted mutation reveals a central role for SR-BI in hepatic selective uptake of high density lipoprotein cholesterol. Proc Natl Acad Sci U S A 1998;95:4619 – 24.
The effect of the APOE polymorphism on HDL-C concentrations depends on the cholesterol ester transfer protein gene variation in a Southern European population Jose´ V. Sorlı´ *, Dolores Corella, Francesc France´s, Judith B. Ramı´rez, Jose´ I. Gonza´lez, Marisa Guille´n, Olga Portole´s Department of Preventive Medicine and Public Health, Genetic and Molecular Epidemiology Unit, School of Medicine, University of Valencia, Avda. Blasco Iban˜ez, 15, 46010 Valencia, Spain Received 5 October 2005; received in revised form 5 October 2005; accepted 5 October 2005 Available online 19 January 2006
Abstract Background: Apolipoprotein E (ApoE) locus has consistently shown a significant association with low-density lipoprotein cholesterol (LDLC). However, its impact on high-density lipoprotein cholesterol (HDL-C) has been highly controversial suggesting that it may be contextdependent. We examined the gene – gene interaction between the common ApoE and the CETP polymorphisms in determining HDL-C concentrations in men and women from the general population. Methods: 550 unrelated Caucasian subjects were randomly selected from a Mediterranean Region in Spain. Plasma lipids, anthropometric, clinical and lifestyle variables were measured. Common ApoE and CETP-TaqIB polymorphisms were determined. Results: We have found a gene – gene interaction between and ApoE and the CETP loci in determining HDL-C concentrations. Thus, after adjustment for gender, age, body mass index, tobacco smoking, alcohol consumption, physical exercise and medication, carriers of the E4 allele had lower HDL-C concentrations [mean and (standard error): 40.1 (2.6) mg/dL] than E2 subjects [47.7 (3.2) mg/dL; p = 0.019], and even lower than those of the E3 subjects [44.7 (1.4) mg/dL; p = 0.042], only if they had the B1B1 genotype. However, mean HDL-C concentrations were higher among those with E4 allele carrying the B2 allele at the CETP gene locus [50.5 (2.3) mg/dL], and lower among E2 subjects carrying the B2 allele [45.5 (2.6) mg/dL]. This interaction was observed in both men and women. This gene – gene interaction remained statistically significant even after additional adjustment for triglycerides. Conclusions: The effect of the ApoE polymorphism on HDL-C concentrations depends on the CETP polymorphism, explaining some of the controversial results previously reported for this polymorphism. D 2005 Elsevier B.V. All rights reserved. Keywords: ApoE; CETP; Polymorphism; Gene – gene interaction; HDL-C
1. Introduction Understanding the relationship between relevant DNA polymorphisms and plasma lipid concentrations is expected to improve the diagnosis, treatment and prevention of cardiovascular diseases (CVD) [1]. Until recently, most of the association studies have focused on elucidat-
* Corresponding author. Tel.: +34 96 3864417; fax: +34 96 3864166. E-mail address: [email protected] (J.V. Sorlı´). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2005.10.001
ing the role of only one or a few polymorphisms in one gene at a time. There is increasing evidence that indicates the genetic polymorphisms contributing to CVD risk do not have the same effects in different environments, or in different genetic backgrounds [2– 5]. In general, these context-dependent genetic effects may involve the socalled gene – environment interactions as well as the gene– gene interactions or epistasis [6,7]. Most previous studies of the genetics of plasma lipid concentrations fail to take context-dependent effects into account. In this regard, there is a growing awareness that the failure to
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
replicate single-locus association studies for lipid metabolism may be due to the fact that gene –environment and gene – gene interactions are the norm rather than the exception [3,7]. Moreover, the few studies analyzing context-dependent effects on lipid traits have focused on environmental factors and little is known about gene – gene interactions [8 –10]. The apolipoprotein E (ApoE) gene (ApoE) is one of the most well studied genes to date. ApoE mediates the catabolism of chylomicron and very low density lipoprotein (VLDL) remnants via a remnant receptor, and the binding of chylomicron remnants, VLDL, and intermediate density lipoproteins to the low density lipoprotein (LDL) receptor [11]. A common three-allele (E2/E3/E4) polymorphism in the ApoE gene, identified in more than 100 different populations worldwide, is potentially one of the most important genetic predictors of plasma LDL-cholesterol (LDL-C) concentrations [11 –15]. The well-documented pattern of increasing LDL-C concentrations from E2 to E3 to E4 across different populations, has been called the invariant effect of this polymorphism [16]. In contrast, the associations between the ApoE polymorphism and plasma levels of high-density lipoprotein (HDL) cholesterol (HDL-C), have been controversial [15,17 – 19], suggesting strong context dependency. Recently, Djousse et al. [20] have reported a gene – environment interaction between the ApoE polymorphism and alcohol intake on plasma HDL-C concentrations in men and women who participated in the National Heart, Lung, and Blood Institute Family Heart Study. However, gene – gene interactions were not examined in this study. In examining gene –gene interactions in determining HDL-C concentrations, the crucial challenge is to select another relevant gene that is functionally linked to the former in the corresponding metabolic pathway and, for which, biological plausibility of functional interactions exit. In this regard, the enzyme cholesteryl ester transfer protein (CETP) plays a key role in the reverse cholesterol transport; the transport of cholesterol from cells of the arterial wall to the liver, which is mediated by ApoE-containing HDL particles [21,22]. ApoE enhances the lipid exchange between lipoproteins mediated by CETP, and alterations in CETP activity contributes to changes in HDL composition and size [23,24]. As both ApoE and CETP contribute to the reverse cholesterol transport, the effects of variations at the CETP locus may very well interact with the ApoE genotype. Several polymorphisms have been described in the CETP gene [25]. Among them, the CETP-TaqIB polymorphism is the variant that has been most firmly associated with HDL-C concentrations in several populations [26]. Therefore, our aim was to investigate the possible gene –gene interaction between the ApoE variation and the CETPTaqIB polymorphism in determining HDL-C concentrations in a randomly selected Southern European population from Spain.
197
2. Materials and methods 2.1. Subjects and study design In the present study, 550 unrelated Caucasian subjects (171 men and 379 women) from the general population were enrolled. All subjects (aged 18 –85 years) were randomly recruited from the Valencia Region on the East Mediterranean coast of Spain. The study protocol was approved by the Ethics Committee on Human Research of Valencia University. We present data obtained from the individuals attending the first cross-sectional examination carried out between February 2000 and January 2002, who were selected as previously reported [27]. The random selection and recruitment of the study subjects was carried out at the Primary Healthcare Center in Paiporta (Valencia). All the study participants were volunteers and gave their informed consent. Previously validated questionnaires were distributed at the time of the medical examination, and participants were invited to fill them in. A venous blood sample was collected during the medical check-up [27]. Alcohol consumption was carefully evaluated and drinkers (subjects with any amount of alcohol consumed) and nondrinkers (alcohol consumption = 0) were considered. Physical activity was estimated from questions about regular leisure-time physical sports and subjects were categorized as sedentary (no regular physical exercise), or active (regular physical exercise). 2.2. Anthropometric, clinical and biochemical data Anthropometric measurements were taken using standard techniques: weight with light clothing by digital scales; height without shoes by fixed stadiometer. Body mass index (BMI) was calculated as weight (kg)/height (m2). Blood pressure was taken with a calibrated mercury sphygmomanometer following the WHO-MONICA (World Health Organization-Multi-national monitoring of trends and determinants in cardiovascular disease) protocol [27]. Participants were instructed to fast for at least 12 h before a morning examination. Plasma total cholesterol and triglycerides were determined by a Technicon Chem 1 assay, and HDL-C was measured in the supernatant after precipitation of apolipoprotein B-containing lipoproteins with heparin-manganese chloride as previously described [27]. LDL-C was calculated according to the Friedewald et al. equation [28] for samples with triglyceride concentrations below 400 mg/dL. 2.3. DNA extraction and genotyping Genomic DNA was isolated from white blood cells by phenol –cloroform extraction and ethanol precipitation [27]. DNA samples were subjected to amplification by the polymerase chain reaction (PCR) in an Eppendorf DNA thermal cycler. For the common ApoE polymorphism,
198
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
Table 1 Demographic, biochemical and lifestyle characteristics of the study subjects Men (n = 171) Women (n = 378) p*
Age (years) Weight (kg) Height (m) BMI (kg/m2) Waist (cm) Hip (cm) Waist-to-hip ratio Glucose (mg/dL) Uric acid (mg/dL) Total-C (mg/dL) LDL-C (mg/dL) HDL-C (mg/dL) Triglycerides (mg/dL) SBP (mm Hg) DBP (mm Hg) Current smokers (%) Alcohol users (%) Taking lipid lowering drugs (%) Regular physical exercise (%)
Mean (S.D.)
Mean (S.D.)
45.1 (18.8) 79.0 (13.5) 1.70 (0.08) 27.2 (4.1) 95.6 (11.7) 99.8 (8.3) 0.96 (0.08) 110.7 (35.2) 5.1 (1.3) 206.9 (42.9) 136.3 (35.0) 44.2 (8.8) 154.6 (94.6) 131.6 (19.9) 79.3 (10.8) 38.1 79.6 7.0 18.7
45.9 (17.2) 68.8 (14.2) 1.57 (0.07) 28.2 (5.8) 88.6 (15.2) 101.4 (12.8) 0.87 (0.07) 103.2 (24.6) 3.6 (1.1) 208.3 (41.4) 136.3 (36.9) 52.4 (13.9) 118.8 (72.3) 127.2 (23.3) 77.4 (12.3) 24.6 54.8 9.3 13.3
0.648 <0.001 <0.001 0.113 <0.001 0.219 <0.001 0.005 <0.001 0.739 0.992 <0.001 <0.001 0.081 0.147 0.007 <0.001 0.379 0.121
S.D.: standard deviation, BMI: body mass index, Total-C: total cholesterol, LDL-C: low density lipoprotein cholesterol, HDL-C: high density lipoprotein cholesterol, SBP: systolic blood pressure, DBP: diastolic blood pressure. * p-value obtained in the comparison between men and women: Student’s t-test or chi-squared test.
genotyping was carried out by the method of Hixson and Vernier [29]. The CETP-TaqIB polymorphism was determined as previously indicated [24]. 2.4. Statistical analysis Normal distribution for all continuous variables was checked. Triglycerides, were skewed and this variable was logarithmically transformed to improve normality. To assess mean differences between genders, Student’s t-test was used. Allele frequencies were estimated by gene counting and v 2 tests were used to test differences between observed and expected frequencies, assuming Hardy – Weinberg equilibrium, and to test differences in percentages between men and women. For multiple comparisons of means between genotypes, one-way analysis of variance was performed. Bonferroni test was applied for multiple comparisons. Men and women were analyzed together after checking the homogeneity of the genotype effect. To control for potential confounders, mean values were adjusted for age, gender, BMI, tobacco smoking, alcohol consumption, physical exercise and medication by multivariate models. Multivariate linear regression analysis with dummy variables for categorical and interaction terms was used to test the null hypothesis of no interaction between the CETP and the ApoE polymorphism after adjustment for potential confounders. Homogeneity of allelic effects according to gender was tested by introducing the corresponding second-order interaction term in the regression model. Standard regres-
sion diagnostic procedures were used to ensure the appropriateness of these models. Analyses were done using the Statistical Package of Social Sciences (SPSS, version 11.0) for Windows.
3. Results Participants were genotyped for the ApoE polymorphism. One woman with the E2/E4 genotype was excluded for the subsequent analyses. Demographic, biochemical, and lifestyle characteristics of the 549 study subjects (171 men and 378 women) are presented in Table 1. The mean age was 45.6 years, with no statistical differences between men and women. All of them were also genotyped for the TaqIB polymorphism at the CETP gene locus. Table 2 shows genotypes for the CETP variants according to the ApoE polymorphism. No differences by gender were detected, and men and women were analyzed together. Genotype frequencies did not deviate from the Hardy –Weinberg equilibrium. Subjects with the E2/E2 and E2/E3 genotypes were grouped as E2, subjects E3/E3 were classified as E3, and E3/E4 or E4/ E4 were grouped as E4. Table 3 shows crude (unadjusted) and adjusted means of HDL-C and triglycerides according to the ApoE and the CETP polymorphisms in the study subjects. Individuals with the B1B2 genotype were grouped with those of the B2B2 genotype as B2 carriers. No statistically significant differences of HDL-C concentrations per ApoE groups were found ( P crude = 0.881; P adjusted = 0.851). Triglyceride concentrations were significantly different by the ApoE polymorphism in the crude model, but differences were borderline significant ( p = 0.058) when the adjustment by age, gender, BMI, tobacco smoking, alcohol consumption, physical exercise and lipid lowering medication was accounted for. In addition, taking into account that the difference in plasma triglycerides between the ApoE genotypes may mask any effect of the ApoE polymorphism on HDL-C concentrations, an adjustment for triglycerides was carried out . However, this adjustment did not affect the previous results (Table 3). On the other hand, the CETP polymorphism was significantly associated with HDL-C. Carriers of the B2 allele had higher HDL-C concentrations ( p = 0.034) than B1B1 homozygotes, and this difference remained statistically significant ( p = 0.002) even after adjustment for covariates. No association of the CETP polymorphism with triglyceride concentrations was found. Additionally, when the association between the ApoE and the Table 2 APOE genotype distribution according to the CETP polymorphism CETP
B1/B1 B2 carrier
APOE genotype E2, n (%)
E3, n (%)
E4, n (%)
22 (10.4) 29 (9.4)
160 (75.5) 264 (76.4)
30 (14.1) 44 (14.2)
The APOE genotype distribution was not significantly different among the CETP genotypes ( p = 0.707).
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
199
Table 3 Crude and adjusted means of HDL-C and triglyceride concentrations depending on the APOE and the CETP polymorphism Genotype
HDL-C (mg/dL)
Triglycerides (mg/dL)
Crude (S.D.)
Adjusted* (S.E.)
Crude (S.D.)
Adjusted* (S.E.)
APOE E2 (n = 51) E3 (n = 424) E4 (n = 74) P
49.7 (10.3) 49.8 (13.4) 48.8 (13.1) 0.881
46.2 (2.2) 46.9 (1.2) 46.0 (1.8) 0.851**
157.9 (96.6) 124.1 (73.8) 142.4 (104.3) 0.042
176.3 (14.8) 140.4 (7.8) 155.0 (11.1) 0.058
CETP B1/B1 (n = 212) B2 carrier (n = 337) P
47.9 (10.9) 50.9 (14.2) 0.034
44.3 (1.4) 48.5 (1.2) 0.002
134.9 (92.2) 126.3 (73.6) 0.553
155.7 (8.9) 139.3 (8.0) 0.175
S.D.: standard deviation, S.E.: standard error. P: P-value obtained in the global comparison between genotypes (ANOVA and ANCOVA tests). * Mean value adjusted for: gender, age, BMI, medical treatment, tobacco smoking, alcohol consumption, and physical activity. ** After additional adjustment for triglycerides, the adjusted P value was non-significant ( P = 0.840) and the adjusted mean values were: 48.4 (1.9); 47.5 (0.8) and 48.2 (1.6) for E2, E3 and E4 genotypes, respectively.
CETP polymorphisms with HDL-C and triglycerides was examined, after excluding those subjects taking lipid lowering drugs (n = 12 men and 35 women), no differences in the statistical significance of the results were found. In addition, no heterogeneity of the effect of the ApoE polymorphism or the CETP polymorphism on these lipid traits by gender was found ( p > 0.05 for all the four gene – gender interactions). Furthermore, the gene –gene interaction between and ApoE and the CETP loci in determining HDL-C and triglyceride concentrations was investigated. For HDL-C concentrations, a statistically significant interaction term was found ( p = 0.047) after adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and lipid lowering medication. The effect of the ApoE polymorphism on HDL-C concentrations depends on the CETP polymorphism. Thus, carriers of the E4 allele had lower HDL-C concentrations [mean and (standard error): 40.1 (2.6) mg/dL] than E2 subjects [47.7 (3.2) mg/dL; p = 0.019], and even lower than those of the E3 subjects [44.7 (1.4) mg/dL; p = 0.042], only if they had the B1B1 genotype. However, mean HDL-C concentrations were higher among those with E4 allele carrying the B2 allele at the CETP gene locus [50.5 (2.3) mg/dL], and lower among E2 subjects carrying the B2 allele [45.5 (2.6) mg/dL]. No heterogeneity of the ApoE –CETP interaction by gender was found. The statistical significance of the interaction term between ApoE – CETP – gender in the multivariate hierarchical regression model was p = 0.296. Moreover, when the homogeneity of the gene –gene interaction was explored in men and women separately, the direction of the interaction effect was the same in both genders. Because of the small number of men in comparison with women, the ApoE – CETP term was statistically significant in women ( p = 0.035), and borderline significant in men. In addition, as 45% of women were post-menopausal, we tested the heterogeneity of the gene – gene interaction effect by
menopausal status and similar results were found for preand post-menopausal women. We further studied the interaction between the ApoE and CETP polymorphisms in determining HDL-C concentrations after the exclusion of subjects taking lipid lowering drugs. In the multivariate model, neither the statistical significance of the interaction term between the ApoE and the CETP polymorphism ( p = 0.047), nor the magnitude of the effect were modified in this sub-sample, confirming the consistency of the interaction. Moreover, to discard the influence of triglyceride concentrations on this interaction, an additional adjustment for plasma triglycerides was carried out. Fig. 1 shows adjusted means for plasma HDL-C depending on the ApoE and the CETP polymorphisms after adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and triglycerides in subjects who were not taking lipid lowering drugs. Finally, the ApoE – CETP interaction in determining triglyceride concentrations was assessed (results not shown). No statistically significant interaction between the ApoE and the CETP polymorphism was observed ( p = 0.590). However, a decrease in the triglyceride concentrations was observed in E4 subjects carrying the B2 allele as compared with the other groups (140.1 T13.6 mg/dL in E4 subjects carrying the B2 allele versus 179.1 T16.1 mg/dL in E4 subjects with the B1B1 genotype; p = 0.247). No heterogeneity of the ApoE – CETP interaction by gender was found in determining triglyceride concentrations ( p = 0.828 for the term ApoE –CETP – gender in the multivariate hierarchical regression model). Although the decrease in triglyceride concentrations in E4 subjects depending on the CETP polymorphism did not reach statistical significance, the concomitant increase of the HDL-C concentrations in carriers of the B2 allele made the triglycerides to HDL-C ratio reach statistical significance in this group. Thus, E4 subjects carrying the B2 allele had a triglycerides to HDL-C ratio statistically lower than E4 homozygous for the B1
200
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
56
P for interaction
54
b
APOE*CETP=0.04 APOE*CETP=0.049
HDL-C (mg/dL) H
52 50
a
48 46
A APOE genotype 44
E2 a b
42
E3 E4
40 B1/B1
B2 carrier
CETP polymorphism Fig. 1. Adjusted means for plasma HDL-C depending on the ApoE and the CETP polymorphisms in men (n = 159) and women (n = 343) who were not taking lipid lowering medications. Mean values were adjustment for gender, age, BMI, tobacco smoking, alcohol consumption, physical exercise and triglycerides. P value for the interaction term was estimated from the multivariate regression model. P values for the comparison between two groups were obtained after multivariate adjustment and correction for multiple comparisons (Bonferroni test). Values marked with the same letter are statistically ( P < 0.05) different. Error bars: standard error of means.
allele: 3.1 (95% CI: 2.1 –4.1) versus 5.3 (95% CI: 4.2– 6.4), respectively.
4. Discussion In this study, we have found a statistically significant gene – gene interaction between the common ApoE polymorphism and the CETP-TaqIB polymorphism in determining plasma HDL-C concentrations in a random sample of Caucasian adults from the general population. This finding may help to explain the controversial results previously reported by dozens of researchers concerning the association between the ApoE polymorphism and HDLC. In contrast to the well-documented association between the common ApoE polymorphism and LDL-C concentrations [11], the association between this polymorphism and HDL-C concentrations has been highly inconsistent [12 – 19]. Thus, although in the majority part of studies no statistically significant association between the common ApoE polymorphism and HDL-C concentrations was observed [12,14,17,30], there were some investigations reporting that HDL-C concentrations tended to be statistically lower in E4 and higher in E2 carriers [15,18,31]. In other studies, only the significant association between the E4 allele and lower HDL-C concentrations was found [32,33]. Moreover, in a review of the ApoE protein, Mahley and Rall [34] reported that, although the effect of the ApoE variation on HDL-C was not significant, a pattern of
increasing HDL-C concentrations from E2 to E3 to E4 alleles can be observed by combining several studies. In the present study, we did not find any significant association between the ApoE polymorphism and HDL-C concentrations when the effect of this polymorphism was analyzed alone. However, when the gene –gene interaction with the CETP-TaqIB polymorphism was considered, we found statistically significant associations between the ApoE polymorphism and HDL-C depending on the CETP genotypes. Thus, in B1B1 subjects, carriers of the E4 allele had the lowest HDL-C concentrations, whereas carriers of the E2 allele had the highest. However, in subjects carrying the B2 allele, this trend was reversed, and mean HDL-C concentrations were higher among those with the E4 allele. In addition, no heterogeneity of the ApoE –CETP interaction by gender was found, decreasing the likelihood that this result has been obtained by chance. These results are in agreement with Miltiadous et al. [35] who studied 200 normolipidemic individuals from Greece and found that participants carrying the E4 allele, the B1 allele of the CETP-TaqIB polymorphism and the allele T of the ApoA4 gene (A to T polymorphism at site 347) had statistically significant lower HDL-C concentrations compared to those not carrying the above allele combination). Although the physiological mechanism that may explain the observed results needs to be elucidated, there are some interesting features related to these polymorphism that may contribute to this interaction. The TaqIB polymorphism has been associated with CETP activity. Thus, in the Framingham
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
Study male and female carriers of the B2 allele had significantly lower CETP activity than did those homozygotes for the B1 allele, supporting the observation that lower CETP activity is associated with increased HDL-C concentrations [24]. In the present study, before the interaction with the ApoE gene was analyzed, we also found that the B2 allele was associated with higher HDL-C concentrations in line with the majority of studies [26]. The mechanism by which TaqIB polymorphism may affect CETP activity is not known. It is unlikely that this polymorphism located in an intron represents a functional mutation. However this polymorphism may have an effect on the transcriptional regulation of the CETP gene or it could be a marker of a still unknown functional mutation [36]. On the other hand, preliminary experimental studies have shown that ApoE increases CETP activity suggesting that the effects of ApoE on the changes of lipid profile in the HDL fraction are not the direct effect of ApoE on lipid transfer, but mediated by CETP activity in the plasma [37]. Several studies have reported that plasma ApoE concentrations are determined in part by the ApoE polymorphism, thus the E2 allele is associated with higher ApoE concentrations; conversely, the E4 allele is related to lower ApoE concentrations [38,39]. These observations are consistent with the faster rate of catabolism of particles containing ApoE4 compared with those containing ApoE3
201
[40]. In some studies examining the effect of the ApoE genotype on CETP activity, a significant interaction between the change in plasma CETP and the ApoE genotype in response to cholesterol feeding has been reported [41]. Thus, Martin et al. [41] observed that E2 subjects responded to cholesterol feeding with a increase of plasma CETP and lower plasma HDL-C, whereas E4 subjects showed a lower increase of plasma CETP and higher plasma HDL-C concentrations. E3 subjects had an intermediate effect between E2 and E4. This situation is similar to the effect that we observed by ApoE genotype in the case of B2 allele. Moreover, taking into account that it has been demonstrated that HDL particle size in carriers of the B2 allele is higher than in B1B1 subjects [24], and that ApoE is essential for the formation of large HDL particles [42,43], it seems reasonable to hypothesise that carriers of the B2 allele have HDL particles more rich in ApoE than B1B1 homozygotes. Considering that the acquisition of ApoE by HDL enables the lipoproteins to interact with LDL and ApoE receptors [44], we suggest that the number of ApoE in HDL is crucial for the receptor interaction. In B2 subjects, the buoyant and enriched-ApoE HDL allows the efficient interaction with the receptors (Fig. 2), independently of the ApoE-receptor affinity determined by the ApoE genotype [19]. Therefore, in carriers of the B2 allele, HDL is efficiently presented and
Fig. 2. Proposed model of the interaction between ApoE and CETP depending on their specific genetic variants. Panel A shows the situation predicted in individuals carrying the B2 allele at the CETP locus. In these subjects, the buoyant and ApoE-enriched HDL particles allow the efficient interaction between ApoE and the scavenger receptor class B-type I (SR-BI), the stimulatory effect of the different ApoE isoforms on the CETP activity prevailing over the different affinity of the ApoE isoforms for the SR-BI as depicted in the figure. Thus, higher plasma HDL-C concentration can be observed in carriers of the E4 allele compared to the E2. Conversely, for CETP B1 allele homozygous subjects (panel B), the small and impoverished HDL have a limited capability of being presented to SR-BI. In this situation, therefore, the differing affinity of the different ApoE isoforms for the receptor prevails, which is higher in E4 allele carriers than those of E2 or E3.
202
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
attracted by the HDL receptor, scavenger receptor BI [45,46], the stimulatory effect of ApoE on CETP activity prevailing over the ApoE-receptor affinity. On the other hand, in B1B1 subjects the smaller and poor-ApoE HDL particles may limit the capacity to present HDL to the scavenger receptor BI. For this reason, in B1B1 subjects, efficient presentation may depend on the differing affinities between the type of ApoE and the receptor, that was previously concealed because the sufficient amount of apolipoprotein in the B2 situation. In conclusion, we have observed a statistically significant interaction between the ApoE and the CETP-TaqIB polymorphisms in determining plasma HDL-C concentrations in men and women from the general population. This interaction is compatible with the previously reported results obtained in experimental models and may explain some of the inconsistencies regarding the association between the ApoE polymorphism and plasma HDL-C concentrations and even cardiovascular risk. Therefore, this gene – gene interaction should be examined in other population studies to confirm consistency and to explore the role of additional genetic or environmental modulating factors. Acknowledgments This study was supported by grants (FIS PI02/1096) and (G03/160) from the Instituto de Salud Carlos III and grant GR04/043 from the Oficina de Ciencia y Tecnologı´a, Generalitat Valenciana. References [1] Ordovas JM. Cardiovascular disease genetics: a long and winding road. Curr Opin Lipidol 2003;14:47 – 54. [2] Talmud PJ, Humphries SE. Gene: environment interactions and coronary heart disease risk. World Rev Nutr Diet 2004;93: 29 – 40. [3] Hunter DJ. Gene – environment interactions in human diseases. Nat Rev Genet 2005;6(4):287 – 98. [4] Ordovas JM, Corella D, Demissie S, et al. Dietary fat intake determines the effect of a common polymorphism in the hepatic lipase gene promoter on high-density lipoprotein metabolism: evidence of a strong dose effect in this gene – nutrient interaction in the Framingham Study. Circulation 2002;106:2315 – 21. [5] Rontu R, Karhunen PJ, Ilveskoski E, et al. Smoking-dependent association between paraoxonase 1 M/L55 genotype and coronary atherosclerosis in males: an autopsy study. Atherosclerosis 2003;171: 31 – 7. [6] Klos KL, Kardia SL, Hixson JE, et al. Linkage analysis of plasma ApoE in three ethnic groups: multiple genes with context-dependent effects. Ann Hum Genet 2005;69:157 – 67. [7] Xin X, Srinivasan SR, Chen W, Boerwinkle E, Berenson GS. Interaction effect of Serine447Stop variant of the lipoprotein lipase gene and C-514T variant of the hepatic lipase gene on serum triglyceride levels in young adults: the Bogalusa Heart Study. Metabolism 2003;52:1337 – 42. [8] Tiret L. Gene – environment interaction: a central concept in multifactorial diseases. Proc Nutr Soc 2002;61:457 – 63. [9] Ordovas JM, Corella D. Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71 – 118.
[10] Moore JH, Williams SM. New strategies for identifying gene – gene interactions in hypertension. Ann Med 2002;34:88 – 95. [11] Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 1988;8:1 – 21. [12] Schaefer EJ, Lamon-Fava S, Johnson S, et al. Effects of gender and menopausal status on the association of apolipoprotein E phenotype with plasma lipoprotein levels. Results from the Framingham Offspring Study. Arterioscler Thromb 1994;14:1105 – 13. [13] Boer JM, Ehnholm C, Menzel HJ, et al. Interactions between lifestylerelated factors and the ApoE polymorphism on plasma lipids and apolipoproteins. The EARS Study. European Atherosclerosis Research Study. Arterioscler Thromb Vasc Biol 1997;17:1675 – 81. [14] Hsueh WC, Mitchell BD, Hixson JE, Rainwater DL. Effects of the ApoE polymorphism on plasma lipoproteins in Mexican Americans. Ann Epidemiol 2000;10:524 – 31. [15] Tan CE, Tai ES, Tan CS, et al. APOE polymorphism and lipid profile in three ethnic groups in the Singapore population. Atherosclerosis 2003;170:253 – 60. [16] Frikke-Schmidt R. Context-dependent and invariant associations between APOE genotype and levels of lipoproteins and risk of ischemic heart disease: a review. Scand J Clin Lab Invest Suppl 2000;233:3 – 25. [17] Pablos-Mendez A, Mayeux R, Ngai C, Shea S, Berglund L. Association of ApoE polymorphism with plasma lipid levels in a multiethnic elderly population. Arterioscler Thromb Vasc Biol 1997;17:3534 – 41. [18] Kataoka S, Robbins DC, Cowan LD, et al. Apolipoprotein E polymorphism in American Indians and its relation to plasma lipoproteins and diabetes. The Strong Heart Study. Arterioscler Thromb Vasc Biol 1996;16:918 – 25. [19] Frikke-Schmidt R, Nordestgaard BG, Agerholm-Larsen B, Schnohr P, Tybjaerg-Hansen A. Context-dependent and invariant associations between lipids, lipoproteins, and apolipoproteins and apolipoprotein E genotype. J Lipid Res 2000;41:1812 – 22. [20] Djousse L, Pankow JS, Arnett DK, Eckfeldt JH, Myers RH, Ellison RC. Apolipoprotein E polymorphism modifies the alcohol – HDL association observed in the National Heart, Lung, and Blood Institute Family Heart Study. Am J Clin Nutr 2004;80:1639 – 44. [21] de Grooth GJ, Klerkx AH, Stroes ES, Stalenhoef AF, Kastelein JJ, Kuivenhoven JA. A review of CETP and its relation to atherosclerosis. J Lipid Res 2004;45:1967 – 74. [22] Kinoshita M, Arai H, Fukasawa M, et al. Apolipoprotein E enhances lipid exchange between lipoproteins mediated by cholesteryl ester transfer protein. J Lipid Res 1993;34:261 – 8. [23] Korhonen T, Hannuksela ML, Seppanen S, Kervinen K, Kesaniemi MJ, Savolainen MJ. The effect of the apolipoprotein E phenotype on cholesteryl ester transfer protein activity, plasma lipids and apolipoprotein A I levels in hypercholesterolaemic patients on colestipol and lovastatin treatment. Eur J Clin Pharmacol 1999;54:903 – 10. [24] Ordovas JM, Cupples LA, Corella D, et al. Association of cholesteryl ester transfer protein-TaqIB polymorphism with variations in lipoprotein subclasses and coronary heart disease risk: the Framingham study. Arterioscler Thromb Vasc Biol 2000;20:1323 – 9. [25] Boekholdt SM, Thompson JF. Natural genetic variation as a tool in understanding the role of CETP in lipid levels and disease. J Lipid Res 2003;44:1080 – 93. [26] Boekholdt SM, Sacks FM, Jukema JW, et al. Cholesteryl ester transfer protein TaqIB variant, high-density lipoprotein cholesterol levels, cardiovascular risk, and efficacy of pravastatin treatment: individual patient meta-analysis of 13,677 subjects. Circulation 2005;111:278 – 87. [27] Sorli JV, Velert R, Guillen M, et al. Effects of the apolipoprotein E polymorphism on plasma lipid levels and cardiovascular disease risk in a Mediterranean population. Med Clin (Barc) 2002;118: 569 – 74. [28] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma,
J.V. Sorlı´ et al. / Clinica Chimica Acta 366 (2006) 196 – 203
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
without use of the preparative ultracentrifuge. Clin Chem 1972;18: 499 – 502. Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990;31: 545 – 8. Muros M, Rodriguez-Ferrer C. Apolipoprotein E polymorphism influence on lipids, apolipoproteins and Lp(a) in a Spanish population underexpressing ApoE4. Atherosclerosis 1996;121:13 – 21. Tiret L, de Knijff P, Menzel HJ, Ehnholm C, Nicaud V, Havekes LM. ApoE polymorphism and predisposition to coronary heart disease in youths of different European populations. The EARS Study, European Atherosclerosis Research Study. Arterioscler Thromb 1994;14:1617 – 24. Gomez-Coronado D, Alvarez JJ, Entrala A, Olmos JM, Herrera E, Lasuncion MA. Apolipoprotein E polymorphism in men and women from a Spanish population: allele frequencies and influence on plasma lipids and apolipoproteins. Atherosclerosis 1999;147:167 – 76. Dallongeville J, Lussier-Cacan S, Davignon J. Modulation of plasma triglyceride levels by ApoE phenotype: a meta-analysis. J Lipid Res 1992;33:447 – 54. Mahley RW, Rall Jr SC. Apolipoprotein E: far more than a lipid transport protein. Annu Rev Genomics Hum Genet 2000;1: 507 – 37. Miltiadous G, Hatzivassiliou M, Liberopoulos E, et al. Gene polymorphisms affecting HDL-cholesterol levels in the normolipidemic population. Nutr Metab Cardiovasc Dis 2005;15:219 – 24. Thompson JF, Durham LK, Lira ME, Shear C, Milos PM. CETP polymorphisms associated with HDL cholesterol may differ from those associated with cardiovascular disease. Atherosclerosis 2005; 181:45 – 53. Kinoshita M, Arai H, Fukasawa M, et al. Apolipoprotein E enhances lipid exchange between lipoproteins mediated by cholesteryl ester transfer protein. J Lipid Res 1993;34:261 – 8.
203
[38] Schiele F, De Bacquer D, Vincent-Viry M, et al. Apolipoprotein E serum concentration and polymorphism in six European countries: the ApoEurope Project. Atherosclerosis 2000;152:475 – 88. [39] Vincent-Viry M, Schiele F, Gueguen R, Bohnet K, Visvikis S, Siest G. Biological variations and genetic reference values for apolipoprotein E serum concentrations: results from the STANISLAS cohort study. Clin Chem 1998;44:957 – 65. [40] Davignon J, Cohn JS, Mabile L, Bernier L. Apolipoprotein E and atherosclerosis: insights from animal models. Clin Chim Acta 1999;286:115 – 43. [41] Martin LJ, Connelly PW, Nancoo SD, et al. Cholesterol ester transfer protein and high density lipoprotein responses to cholesterol feeding in men: relationship to apolipoprotein E genotype. J Lipid Res 1993; 34:437 – 46. [42] Krimbou L, Marcil M, Chiba H, Genest Jr J. Structural and functional properties of human plasma high density-sized lipoprotein containing only ApoE particles. J Lipid Res 2003;44:884 – 92. [43] Jiang XC, Beyer TP, Li Z, et al. Enlargement of high density lipoprotein in mice via liver receptor activation requires apolipoprotein E and is abolished by cholesteryl ester transfer protein expression. J Biol Chem 2003;278:49072 – 8. [44] Dergunov AD, Smirnova EA, Merched A, et al. Structural peculiarities of the binding of very low density lipoproteins and low density lipoproteins to the LDL receptor in hypertriglyceridemia: role of apolipoprotein E. Biochim Biophys Acta 2000;1484:29 – 40. [45] Arai T, Rinninger F, Varban L, et al. Decreased selective uptake of high density lipoprotein cholesteryl esters in apolipoprotein E knockout mice. Proc Natl Acad Sci U S A 1999;96:12050 – 5. [46] Varban ML, Rinninger F, Wang N, et al. Targeted mutation reveals a central role for SR-BI in hepatic selective uptake of high density lipoprotein cholesterol. Proc Natl Acad Sci U S A 1998;95:4619 – 24.