Fenofibrate Effect on Triglyceride and Postprandial Response of Apolipoprotein A5 Variants

【摘要】  Objective— Apolipoprotein A5 ( APOA5 ) is a key determinant of plasma triglyceride (TG) concentrations. Genetic variation at the APOA5 locus could be responsible for some of the observed differences in response to fenofibrate therapy.

Methods and Results— We examined the association between G) at APOA5 and TG and HDL-C response to fenofibrate and a postprandial lipid challenge in 791 men and women participating in the GOLDN study. After 3-week drug treatment, APOA5 56G carriers displayed significant decrease in TG ( P =0.006), and increase in HDL-C ( P =0.002) levels relative to their basal values in the fasting state when compared with noncarriers (a TG reduction of –35.8±2.8% versus –27.9±0.9% and a HDL-C increase of 11.8±1.3% versus 6.9±0.5%, respectively). In the postprandial lipemia after a fat load, the 56G carriers showed a significant decrease in the area under curve for TG and increase for HDL-C than the noncarriers. These diverse beneficial responses of 56G carriers to fenofibrate were further characterized by a higher increase in large LDL-C concentrations and LDL size. On the other hand, subjects with different APOA5 C genotypes showed no significant response to fenofibrate intervention.

Conclusion— This study suggests that the APOA5 56G carriers benefited more from the fenofibrate treatment than noncarriers in lowering plasma TG and increasing HDL-C levels.

We examined association between variants at APOA5 and TG and HDL-C response to fenofibrate and a postprandial lipid challenge in 791 men and women participating in the GOLDN study, and observed that the 56G carriers benefited more from the fenofibrate treatment than noncarriers in lowering TG and increasing HDL-C levels

【关键词】  APOA fenofibrate triglyceride lowering increasing HDLC genedrug interaction

Introduction

Elevated plasma triglyceride (TG) levels (hypertriglyceridemia), one of the characteristics of the metabolic syndrome (MS), have been identified as an independent risk factor of coronary heart disease (CHD). 1,2 The atherogenic potential of hypertriglyceridemia can be contributed in part by postprandial lipoprotein remnants, 3,4 which can be internalized into the arterial intima and contribute to the formation of foam cells and the atherosclerotic plaque. 5 Therefore, lowering of TG levels can result in CHD risk reduction. 6 Fenofibrate, an agonist of the peroxisome proliferative activated receptor alpha protein (PPARA), has been shown to significantly reduce serum TG levels. 7,8 However, dramatic interindividual variations in response to fenofibrate have been observed, and they could be partially driven by genetic factors. The potential interactions between fenofibrate therapy and common polymorphisms at candidate genes have been barely explored. 9,10 See page 1224

The apolipoprotein A5 locus ( APOA5 ) resides on the long arm of chromosome 11, about 30 kb proximal to the APOA1/APOC3/APOA4 gene cluster. 11 Transgenic mice with human APOA5 ( hAPOA5 ) exhibited significantly reduced TG levels, whereas the APOA5 knockout mice had significantly increased plasma TG concentrations as compared with wild-type mice. 11 Human APOA5 protein has been detected at very low concentrations (24 to 406 µg/L) in plasma as a component of HDL, VLDL, and chylomicron particles. 12,13 Moreover, plasma APOA5 levels have been negatively associated with plasma TG and positively with HDL-C concentration. 12,13 These observations indicate the important roles of APOA5 in lipid metabolism and homeostasis. To determine whether APOA5 sequence variation influences plasma TG levels and risk of CHD in humans, several common single nucleotide polymorphisms (SNPs, C) at the APOA5 locus have been identified. 11 APOA5 G SNP is a nonsynonymous substitution (from serine ) to tryptophan ). This polymorphism was predicted to change the structure and properties of the APOA5 protein. 14,15 However, possible functions of other SNPs remain to be explored. We and others have demonstrated that the minor alleles of these SNPs are significantly associated with increased plasma TG levels in populations of diverse ethnicities. 14,16–20 Furthermore, we have shown that these APOA5 variants are associated with increased VLDL, RLP-TG, and RLP-C concentrations in the Framingham Heart Study. 14 Consistent with these findings, studies examining postprandial lipemia after a fat load have reported significant associations between the APOA5 C SNP and TG-rich lipoproteins (TRL) during the postprandial period. 21,22 Moreover, APOA5 variants have been associated with increased risk for CHD, 14 especially relevant in Asian populations as the minor allele frequencies are notably higher in these populations. 19,23,24 Furthermore, we examined whether dietary fat intake modulates the effect of APOA5 variants on TG levels, and demonstrated a significant interaction between dietary n-6 PUFA intake and the APOA5 C SNP, influencing the levels of TG and RLP. 25 In addition, fenofibrate is known to induce APOA5 expression in human primary hepatocytes; a PPARA response element was identified in the promoter region of APOA5. 26 This suggests that APOA5 is the target of fenofibrate. Together, these findings prompted us to examine the role of APOA5 genotypes in differential response to PPARA agonists. Therefore, we tested the hypothesis that carriers of certain APOA5 variants display differential responses to fenofibrate, before and after ingestion of a high fat meal, with respect to the plasma levels of TG, HDL-C, and subclasses of lipoproteins.

Methods and Materials

Study Design and Subjects

The study sample comprises 393 men and 398 women who participated in the Genetics of Lipid Lowering Drugs and Diet Network (GOLDN) and who have full data record for the pre- and post-drug intervention for the postprandial TG test. Detailed design and methodology for the GOLDN are described in supplement I, available online at http://atvb.ahajournals.org. Briefly, clinical and biochemical measurements were collected in all individuals before and after exposure to the drug fenofibrate. In addition, a postprandial study fat-challenge was also carried out before and after fenofibrate treatment. Participants were given a supply of fenofibrate as 160 mg tablets (TriCor, Abbott Laboratories) and informed to take one fenofibrate tablet with a breakfast meal once daily for 3 weeks.

Laboratory Methods

TGs were measured by glycerol-blanked enzymatic method on the Roche COBAS FARA centrifugal analyzer (Roche Diagnostics Corporation). The GOLDN study measured NMR LDL and HDL particle size in addition to TG-rich lipoproteins and remnant particles. This method uses signal amplitudes of the lipoprotein subclasses of difference sizes as its basis of quantification. 27 Comparison of NMR and ultracentrifugation separation in this study population showed a high degree of correlation, suggesting that NMR is a valid alternative method for measuring TG-rich lipoproteins. 28 Blind duplicate samples from 5% of participants were sent to the laboratory to assess repeatability. For all lipid subfractions, the repeatability was above 90%. All blood samples from each individual were stored until the completion of their participation and then analyzed together.

Genetic Analysis

Genomic DNA was isolated from peripheral blood leukocytes using Puregene DNA reagents following the vendor?s protocol. As APOA5 C (rs662799 G (rs3135506) represent the two tag SNPs within APOA5 locus in the white population, 14,16,17 both were genotyped with Applied Biosystems TaqMan SNP genotyping system. 29

Statistical Analyses

All statistical analyses were performed using SAS 9.1. (Cary). Continuous variables, such as TG and TRL concentrations, that were not normally distributed, were log-transformed to achieve normality before fitting statistical models. We assessed the relationship between APOA5 genotypes, drug, and plasma lipid phenotypes by covariance analysis. The familial relations within the population were adjusted using a generalized linear model implemented in the GENMOD procedure in SAS assuming an exchangeable correlation structure within pedigree. 14,30

To examine the postprandial TG and HDL-C response of APOA5 genotypes to drug intervention across 3 time points, we fitted a 3-level and individual growth mixed model using SAS Proc Mixed 30,31 : level 1, individual measurements across three time points; level 2, individual nested within pedigree; and level 3, pedigree. To model the individual growth curve, we used an autoregressive AR(1) error covariance matrix, while treating the intercept and time as random effects. The individuals nested within pedigrees were modeled using a generalized linear model. 14,30 We also measured the postprandial response by computing the area under curve (AUC), which was defined as the area between the plasma concentration of the corresponding parameter versus time curve and a line drawn parallel to the horizontal axis through the 0 hour concentration and for which we used the generalized estimating equations approach 14,30 as implemented in the GENMOD procedure in SAS. Meanwhile, lipid response phenotypes (the dependent variable) were adjusted for potential confounders including gender, age, BMI, smoking, alcohol consumption, physical activity, postmenopausal status, hormone use, drugs for lowering cholesterol, diabetes, and hypertension. Men and women were analyzed together, as well as separately to examine gender specific effects. The percent change for each parameter caused by fenofibrate treatment was calculated by dividing the difference x 2– x 1 (subtracting the initial value from the final value), by the initial value, and multiplying by 100. Probability values <0.05 were considered statistically significant.

Results

Demographic, Biochemical, and Genotypic Characteristics of Study Participants

Demographic, biochemical, and genotypic characteristics of the participants at baseline are presented in Table 1. Genotypes of both APOA5 C, were in Hardy–Weinberg equilibrium. Consistent with previous studies, 14,25 these two variants were not in linkage disequilibrium among unrelated subjects ( P =0.745), thus each polymorphism represents an independent variant at the APOA5 locus. 14,16,17 The frequencies C are 5.6% and 5.7%, respectively. No significant difference in genotype frequency was observed between men and women.

TABLE 1. Demographic, Biochemical, and Genotypic Characteristics of Participants at the Baseline According to Sex

APOA5 Genotypes Modulate Plasma Lipid Response to Fenofibrate Treatment

To determine how subjects with various APOA5 genotypes respond to fenofibrate treatment, we examined differences in fasting plasma lipid levels at baseline and after drug intervention among carriers and noncarriers of APOA5 variants. As we did not observe any significant interaction between gender and genotype, men and women were combined for subsequent analyses. At baseline (see Table 2 ), the minor allele 56G carriers (ie, GG+CG) exhibited significantly higher plasma TG (176±137 mg/dL versus 135±91 mg/dL; P =0.012) and lower HDL-C levels ( P =0.026) than noncarriers (CC). However, after 3 weeks of drug intervention, the statistical significance of the differences between these 2 groups disappeared for plasma TG (94±47 mg/dL versus 89±55 mg/dL; P =0.220) and HDL-C ( P =0.715) concentrations. Such differential changes in TG and HDL-C concentrations between 2 groups from pre-drug to post-drug in response to fenofibrate treatment are displayed in Figure 1A and 1 C, respectively. To further illustrate the differential response of 56G carriers to fenofibrate, changes (ie, response) in both TG and HDL-C levels before and after drug administration were calculated and were then fit to the same statistical model. We observed a consistent and differential response between 56G carriers and noncarriers for TG ( P =0.006) and HDL-C ( P =0.002) concentrations (see Figure 1A and 1 C, and Table 2 ). This significance persisted even after adjustment for the corresponding fasting TG and HDL-C levels at baseline ( P =0.032 and 0.007, respectively). Expressed as percent of change, and after adjustment for all the covariates, we observed a statistically significant ( P =0.035) reduction in TG concentrations between carriers of the minor 56G allele and CC subjects (–35.8±2.8% versus –27.9±0.9%, respectively). Similarly, increases in HDL-C were significantly higher ( P =0.008) in carriers of the minor 56G allele than in noncarriers (+11.8±1.3% versus –6.9±0.5%, respectively). Moreover, the differential TG response of 56G carriers was independent of the HDL-C change, as the TG response of 56G carriers to drug was still significantly different from that of noncarriers ( P =0.037) after adjustment for the HDL-C effect. However, the APOA5 G variant showed no significant effect on levels of LDL-C, total cholesterol, and CRP both at baseline and after drug intervention ( Table 2 ).

TABLE 2. Lipid Response to Fenofibrate According to APOA5 G Genotypes

Figure 1. Changes in TG and HDL-C concentrations (mg/dL) in responses to fenofibrate according to APOA5 genotypes. A and B, Plasma TG changes from before-drug to after-drug are shown for APOA5 C (B): –1131C carriers (long dash line=TC +CC) and noncarriers (gray line=TT). C and D, Plasma HDL-C changes from before-drug to after-drug are shown for APOA5 C (D): –1131C carriers (long dash line=TC +CC) and noncarriers (gray line=TT). TG and HDL-C concentrations were adjusted for family relationships, sex, age, BMI, smoking, alcohol use, physical activity, postmenopausal status, hormone use, drugs for lowering cholesterol, diabetes, and hypertension (see Table 2 and supplemental Table I). Probability values were obtained in the multivariate models in predicting the changes in TG and HDL-C for the corresponding APOA5 SNPs.

At baseline, the minor allele –1131C carriers (TC+CC) had also significantly higher TG levels (171±130 mg/dL versus 136±92 mg/dL; P =0.011), but no significant difference for HDL-C levels ( P =0.096) than noncarriers (TT; see supplemental Table I, available online at http://atvb.ahajournals.org). After drug intervention, both –1131C carriers and noncarriers showed no significant differences for TG levels ( P =0.062) or HDL-C ( P =0.420) levels. Such responses to fenofibrate treatment in TG and HDL-C concentrations between 2 groups are shown in Figure 1B and 1 D. In addition, as shown in the response analysis to drug (supplemental Table I), these 2 groups displayed no significant difference for TG ( P =0.114) and HDL-C ( P =0.081) responses. As example, expressed in percent of change, no statistically significant differences in the decrease of TG were observed between –1131C carriers and non-carriers (–31.7±2.7% versus –28.4±0.9%, respectively; P =0.241). Similar to the APOA5 G variant, –1131C carriers and noncarriers showed no difference in the response to drug treatment for LDL-C, total cholesterol and CRP levels.

Moreover, to separate the potential influence of high TG concentrations at baseline on the differential response by genotype, we carried out an additional analysis by classifying subjects into three groups. As 175 mg/dL was the mean of TG in subjects carrying G SNPs, this cut-off point was considered. In this sub-sample of subjects 175 mg/dL, the following groups were considered: (1) subjects carrying the 56G allele (n=34); (2) Subjects carrying the –1131C allele (n=29); (3) subjects without the 56G and –1131C allele (n=131). After multivariate adjustment, statistically significant differences in response among these groups was found ( P =0.010). After fenofibrate treatment, the decrease in TG was statistically higher in carriers of the 56G allele (–53.3±2.4%) when compared with carriers of the –1131C allele (–42.2±3.9%; P =0.021) or with subjects without the 56G and –1131C allele (39.9±1.8%; P =0.010). No statistically significant differences in the decrease of TG was found between carriers of the –1131C allele and subjects without the 56G and –1131C allele ( P =0.550).

Postprandial Response of TG and HDL-C Levels to Fenofibrate Influenced by APOA5 Genotypes

To further understand the nature of the differential response to fenofibrate intervention exhibited by carriers of APOA5 variants, we examined postprandial TG and HDL-C responses of subjects with various APOA5 genotypes to drug intervention after a high fat challenge.

At baseline, the APOA5 G SNP showed significant effects on the postprandial response to a fat load ( Figure 2 ). The 56G carriers had significantly higher TG ( P =0.003) across all 3 time points when compared with noncarriers. In contrast, after drug intervention the TG of 56G carriers are no longer different from those of noncarriers ( P =0.129). These findings were further illustrated by calculating the AUC for postprandial TG. Before fenofibrate treatment the AUC of TG was greater in carriers of the 56G allele when compared with homozygotes for the allele C. Thus, after adjustment for family relationships, gender, age, BMI, smoking, alcohol consumption, physical activity, menopausal status, hormone use, drugs for lowering cholesterol, diabetes, and hypertension, we found statistical significant differences between the estimated area under curve (in arbitrary units) between 56G carriers (2107±138) and noncarriers (1789±115; P =0.025). After fenobifrate treatment, the difference in AUC for TG between 56G carriers and noncarriers was no longer significant (1255±84 versus 1183±70; P =0.246). Further adjustment of postprandial curves for baseline TG resulted in nonsignificant genotype effects in both before and after drug fat loads ( P =0.651 and P =0.256, respectively). In terms of HDL-C, carriers for the 56G allele have a lower postprandial response ( P =0.030 in the mean AUC comparison in the multivariate adjusted model) than subjects homozygous for the T allele before the fenobifrate intervention. After drug intervention, the difference in AUC was no longer significant ( P =0.615). Further adjustment of postprandial curves for baseline HDL-C resulted in non-significant genotype effects.

Figure 2. Postprandial TG and HDL-C response (mg/dL) to a fat-load according to APOA5 56C G genotypes before and after fenofibrate intervention. TG and HDL-C concentrations were adjusted for family relationships, sex, age, BMI, smoking, alcohol use, physical activity, postmenopausal status, hormone use, drugs for lowering cholesterol, diabetes, and hypertension. A and B, Postprandial TG response before and after fenofibrate, respectively, according to APOA5 G genotypes (dash line=CG+GG, solid line=CC). C and D, Postprandial HDL-C response before and after fenofibrate, respectively, according to APOA5 G genotypes (dash line=CG+GG, solid line=CC). Probability values indicate the multivariate adjusted genotype effects in the postprandial period. After additional adjustment for baseline concentrations of TG or HDL-C, these probability values did not remain statistically significant.

The postprandial response of subjects with APOA5 C genotypes is presented in Figure 3. At baseline (ie, before fenofibrate treatment), the minor allele –1131C carriers showed significantly higher TG levels after fat load across all 3 time points compared with the noncarriers ( P =0.002). For HDL-C levels, however, no significant difference was observed between –1131C carriers and noncarriers ( P =0.425). After drug intervention, the –1131C carriers still showed significant differences for TG ( P =0.011) and nonsignificant differences for HDL-C levels ( P =0.515). Thus, as illustrated in Figure 3, the response of –1131C carriers to fenofibrate did not reach statistical significance although a trend in reducing TG levels, similar to that of APOA5 56G carriers, was observed. Further adjustment for baseline triglycerides resulted in non-significant genotype effects.

Figure 3. Postprandial TG and HDL-C response (mg/dL) to a fat-load according to APOA5 C genotypes before and after fenofibrate intervention. TG and HDL-C concentrations were adjusted for family relationships, sex, age, BMI, smoking, alcohol use, physical activity, postmenopausal status, hormone use, drugs for lowering cholesterol, diabetes, and hypertension. A and B, Postprandial TG response before and after fenofibrate, respectively, according to APOA5 C genotypes (dash line=TC+CC, solid line=TT). C and D, Postprandial HDL-C response before and after fenofibrate, respectively, according to APOA5 C genotypes (dash line=TC+CC, solid line=TT). Probability values indicate the multivariate adjusted genotype effects in the postprandial period. After additional adjustment for baseline concentrations of TG or HDL-C, these probability values did not remain statistically significant.

APOA5 Variants Influence Responses of Lipoproteins Subclasses to Fenofibrate

To examine how fenofibrate affects the distribution of lipoprotein subclasses, we analyzed the changes of lipoprotein subclass levels among subjects with various APOA5 genotypes in the response to drug treatment. The results for levels of VLDL, HDL-C, LDL-C lipoprotein subclasses, and their respective particle sizes are presented in Table 3 and supplemental Table II according to APOA5 genotypes.

TABLE 3. Fenofibrate Effects on Sizes of Lipoprotein According to APOA5 Genotypes

At baseline, 56G carriers showed significantly higher levels of total VLDL ( P =0.004) and medium VLDL ( P =0.001), but no significant difference for large VLDL ( P =0.073) and small VLDL ( P =0.436) relative to carriers of the common 56C allele (see supplemental Table II). Fenofibrate intervention significantly reduced total VLDL concentration of the 56G carriers from 2.197 (at baseline, log10-transformed; before transformation, the unit was mg/dL) to 1.857 (after drug, log10-transformed), whereas the 56G noncarriers changed from 2.081 to 1.817 (log10-transformed). Consistent with these observations, the response difference between 56G carriers and noncarriers for total VLDL concentration to fenofibrate was significant ( P =0.016). For 56G carriers, however, it is not apparent which particular subclass of VLDLs was most responsive to the drug.

For HDL-C and LDL-C subclasses, the carriers of the minor 56G allele responded to fenofibrate differently than noncarriers (see supplemental Table II). At baseline, 56G carriers had significantly lower levels of both large HDL-C and large LDL-C ( P =0.007 and 0.016) than noncarriers. After drug intervention, 56G carriers exhibited levels similar to noncarriers ( P =0.453 and 0.791). Thus, as compared with pretreatment values, the 56G carriers produce a significant increase in both large HDL-C and large LDL-C particles after fenofibrate treatment ( P =0.015 and 0.024) compared with noncarriers. After adjustment for baseline values, these differences were not significant for large HDL-C ( P =0.115) and for large LDL-C particles ( P =0.609). Furthermore, the effect of fenofibrate on 56G carriers for concentrations of medium HDL-C, small LDL-C and medium LDL-C was not significant.

We next analyzed VLDL, HDL, LDL particle size changes before and after fenofibrate intervention according to APOA5 genotypes ( Table 3 ). At baseline, 56G carriers had significant smaller LDL than noncarriers ( P =0.008). After drug treatment, the LDL size was similar in carriers and noncarriers ( P =0.717) because of a more pronounced and significant drug-related increase in LDL size of 56G carriers ( P =0.047). Expressed as percent of change, fenofibrate treatment resulted in a higher increase of LDL particle diameter in carriers of the 56G allele than in noncarriers (1.7±0.44% versus 0.6±0.14%; P =0.013, G genotypes and drug was detected in relation to changes in HDL and VLDL sizes ( P =0.497 and 0.906, respectively).

At baseline, the APOA5 –1131C allele was significantly associated with higher concentrations of IDL ( P =0.038), total VLDL ( P =0.015), and intermediate VLDL ( P =0.032). However, –1131C carriers and noncarriers responded similarly to fenofibrate treatment for all the lipoprotein subfractions examined (see supplemental Table II). Furthermore, no significant interaction C SNP in relation to changes in VLDL, HDL, and LDL sizes ( Table 3 ) in the response to drug treatment.

Discussion

This study demonstrates that subjects who carry different APOA5 variants responded differentially to fenofibrate intervention with respect to modifying TG and HDL-C levels. After drug treatment, in the fasting state APOA5 56G carriers displayed a statistically significant higher decrease in TG and higher increase in HDL-C levels relative to their values before the drug when compared with the noncarriers. During the postprandial phase, before fenofibrate intervention, the 56G carriers had higher TG and lower HDL-C levels than the noncarriers across the 3 time points of the postprandial intervention. After fenofibrate treatment, the genotype-specific differences in postprandial response disappeared. Such differential response in the postprandial state between G genotypes was also demonstrated in AUC of TG and HDL-C concentrations. These diverse beneficial responses to fenofibrate observed in 56G carriers were further exhibited as an increase in large LDL-C concentration and LDL size, and as a decrease in total VLDL levels in the fasting state. In contrast, the 56G carriers did not change significantly relative to noncarriers in LDL or total cholesterol, chylomicron, and CRP levels in response to fenofibrate. On the other hand, the differential responsiveness of subjects with APOA5 C genotypes was not associated with lipid values in response to the fenofibrate intervention. We found that –1131C carriers had no statistically significant change relative to noncarriers in plasma levels of TG, HDL-C, or their subclasses after drug intervention, either in the fasting state or in the postprandial response. This suggests that subjects with APOA5 C genotypes react to G genotypes.

The APOA5 G SNP is a nonsynonymous substitution (from serine to tryptophan ), which was predicted to change the structure of the APOA5 protein and potentially to alter the efficiency of either insertion of the nascent polypeptide chain into the endoplasmic reticulum lumen or cleavage of the signal peptide, or by altered lipid affinity. 14,15 This prediction is supported by the recent finding that APOA5 might act as a receptor ligand (guide or bridge) for VLDL and chylomicron to proteoglycan-bound LPL for lipolysis. 32 In addition, based on a signal peptide secretory alkaline phosphatase fusion protein assay, HepG2 cells transfected with the Trp-19 (56G) construct secreted alkaline phosphatase at about 50% the level of cells expressing the common Ser-19 (56C) construct, demonstrating G is indeed a functional variant. 15 Assuming the 19W APOA5 (56G) functions half as effectively as the common 19S APOA5 (56C), one may ask why the 56G carriers could respond more effectively to fenofibrate than noncarriers. Fenofibrate highly upregulates APOA5 expression through PPARA-RXR dimer binding, 26 and this increased expression could compensate for the structural, functional defect of the 19W APOA5. In other words, the effect of 19W APOA5 is normalized on fenofibrate intervention and 56G carriers respond as effectively as non-carriers in clearance of plasma TG and in response to a fat load.

Recent findings showed that elevated plasma TG levels in humans were positively correlated with APOA5 levels in APOA5 56G carriers. 13,33 Thus, the greater response of 56G carriers to fenofibrate would not be explained by the increase of APOA5 expression. An alternative hypothesis is that 56G carriers have a dysfunctional APOA5 protein, which might act as a receptor ligand for VLDL and chylomicron particles that are targeted to proteoglycan-bound LPL for lipolysis. 32,34 At baseline the defective 19W APOA5 protein might act less effectively as a receptor ligand, or signal peptide, than the wild-type 19S APOA5 protein. As a result, the process of APOA5-guided lipolysis through proteoglycan-bound LPL is delayed or attenuated. Thus, turnover rates of APOA5 and TRL are lower in 56G carriers than noncarriers. Consistent with this notion, we and others have demonstrated that 56G carriers had higher levels of plasma TG and VLDL. 14,16–21 This hypothesis is supported by observations in this study that 56G carriers had higher total VLDL levels ( P =0.006), but no significant difference in VLDL size ( P =0.451) than noncarriers. After fenofibrate intervention, by an unknown mechanism, fenofibrate may mediate the exchange of TG/cholesterol between HDL-C and VLDL, leading to reduction of TG contained by VLDL or other TRL. Therefore, under the condition of fenofibrate-mediated reduction of TG levels, the 19W APOA5 can function as effectively as 19S APOA5. Subsequently, after fenofibrate intervention 56G carriers have similar levels of TG and HDL-C as noncarriers. This interpretation is in agreement with the observation in this study that 56G carriers and noncarriers exhibited similar levels of total VLDL after drug intervention ( P =0.202). In addition, 56G carriers had a greater change in LDL size after fenofibrate intervention than did noncarriers, but both groups displayed no significant changes in VLDL and HDL size either at baseline or after drug treatment. This observation is quite consistent with the effect of fibrates on LDL size. 35

The observation that APOA5 G reacted differentially to fenofibrate was not surprising. Each of these SNPs represents an independent haplotype. 14,16,17 Both are associated with TG levels, but display differential associations with HDL-C levels. 14,25 Evidently, the 56G carriers have higher level of plasma APOA5 protein than noncarrier. 33 Conversely, the –1131C carriers has lower APOA5 protein levels than noncarriers. 13,33 In addition, C genotypes, G genotypes, by influencing TG and TRL levels. 25 This further illustrates that the 2 variants respond differentially to nutritional factors, suggesting they could represent either pleiotrophic effects of APOA5 protein, or 2 different proteins in the APOA1/C3/A4/A5 G SNP appears to be functional, 15 thus likely it presents APOA5 variant. However, C SNP. 15 C was shown to be in strong linkage disequilibrium with the functional variant APOC3 T, which affects APOC3 normal response to insulin. 36 Thus, it could represent other genes within the A1/C3/A4/A5 C could be a functional variant of APOA5 by itself, which responds to other stimulants such as n-6 PUFA, 25 instead of fenofibrate. Thus, the function C remains to be explored.

In summary, we report allele-specific responses to fenofibrate intervention. At the APOA5 G genotypes C genotypes. Specifically, the TG lowering and HDL-C raising effects of fenofibrate among carriers of the minor allele APOA5 56G are greater than in noncarriers such that significant differences between these groups observed prior to treatment are essentially attenuated by a 3-week regimen of fenofibrate therapy.

Acknowledgments

We want to thank Dr Ming Ji for advice on statistical analysis, and Jian Shen for discussion and technical assistance. We thank the families for their participation in this research.

Sources of Funding

This study was supported by contract 53-K06-5-10 from NIH and 58-1950-9-001 from the US Department of Agriculture Research Service, and by NIH Heart, Lung, and Blood Institute grant U 01 HL72524, Genetic and Environmental Determinants of Triglycerides, and grant CB06/03/0035 from the ISCIII, Spain. We acknowledge Abbott Laboratories (Abbott Park, Ill) for their supply of study medication for this project.

Disclosures

None.

【参考文献】
  Fruchart JC, Nierman MC, Stroes ESG, Kastelein JJP, Duriez P. New risk factors for atherosclerosis and patient risk assessment. Circulation. 2004; 109: (Suppl. III) III-15–III-19.

Szapary PO, Rader D. The triglyceride-high-density lipoprotein axis: an important target of therapy? Am Heart J. 2004; 148: 211–221.

Zilversmit DB. Atherogenesis: a postprandial phenomenon. Circulation. 1979; 60: 473–485.

Karpe F. Postprandial lipoprotein metabolism and atherosclerosis. J Intern Med. 1999; 246: 341–355.

Gianturco SH, Lin AH, Hwang SL, Young J, Brown SA, Via DP, Bradley WA. Distinct murine macrophage receptor pathway for human triglyceride-rich lipoproteins. J Clin Invest. 1988; 82: 1633–1643.

Havel RJ. Remnant lipoproteins as therapeutic targets. Curr Opin Lipidol. 2000; 11: 615–620.

Staels B, Dallongeville J, Auwerx J, Schoonjans K, Leitersdorf E, Fruchart JC. Mechanism of action of fibrates on lipid and lipoprotein metabolism. Circulation. 1998; 98 (19): 2088–93.

Schoonjans K, Staels B, Auwerx J. Role of the peroxisome proliferators activated receptor (PPAR) in mediating effects of fibrates and fatty acids on gene expression. J Lipid Res. 1996; 37: 907–925.

Christidis DS, Liberopoulos EN, Kakafika AI, Miltiadous GA, Cariolou M, Ganotakis ES, Mikhailidis DP, Elisaf MS. The effect of apolipoprotein E polymorphism on the response to lipid-lowering treatment with atorvastatin or fenofibrate. J Cardiovasc Pharmacol Ther. 2006; 11 (3): 211–21.

Brisson D, Ledoux K, Bosse Y, St-Pierre J, Julien P, Perron P, Hudson TJ, Vohl MC, Gaudet D. Effect of apolipoprotein E, peroxisome proliferator-activated receptor alpha and lipoprotein lipase gene mutations on the ability of fenofibrate to improve lipid profiles and reach clinical guideline targets among hypertriglyceridemic patients. Pharmacogenetics. 2002; 12 (4): 313–20.

Pennacchio LA, Olivier M, Hubacek JA, Cohen JC, Cox DR, Fruchart JC, Krauss RM, and Rubin EM. An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science. 2001; 294: 169–173.

O?Brien PJ, Alborn WE, Sloan JH, Ulmer M, Boodhoo A, Knierman MD, Schultze AE, Konrad RJ. The novel apolipoprotein A5 is present in human serum, is associated with VLDL, HDL, and chylomicrons, and circulates at very low. Clin Chem. 2005; 51 (2): 351–9.

Ishihara M, Kujiraoka T, Iwasaki T, Nagano M, Takano M, Ishii J, Tsuji M, Ide H, Miller IP, Miller NE, Hattori H. A sandwich enzyme-linked immunosorbent assay for human plasma apolipoprotein A-V concentration. J Lipid Res. 2005; 46 (9): 2015–22.

Lai CQ, Demissie S, Cupples LA, Zhu Y, Adiconis X, Parnell LD, Corella D, Ordovas JM. Influence of the APOA5 locus on plasma triglyceride, lipoprotein subclasses, and CVD risk in the Framingham Heart Study. J Lipid Res. 2004; 45: 2096–2105.

Talmud PJ, Palmen J, Putt W, Lins L, Humphries SE. Determination of the functionality of common APOA5 polymorphisms. J Biol Chem. 2005; 280 (31): 28215–20.

Pennacchio LA, Olivier M, Hubacek JA, Krauss RM, Rubin EM, and Cohen JC. Two independent apolipoprotein A5 haplotypes influence human plasma triglyceride levels. Hum Mol Genet. 2002; 11 (24): 3031–3038.

Talmud PJ, Hawe E, Martin S, Olivier M, Miller GJ, Rubin EM, Pennacchio LA, and Humphries SE. Relative contribution of variation within the APOC3/A4/A5 gene cluster in determining plasma triglycerides. Hum Mol Genet. 2002; 11 (24): 3039–3046.

Nabika T, Nasreen S, Kobayashi S, and Masuda J. The genetic effect of the apoprotein AV gene on the serum triglyceride level in Japanese. Atherosclerosis. 2002; 165: 201–204.

Lai CQ, Tai ES, Tan CE, Cutter J, Chew SK, Zhu YP, Adiconis X, Ordovas JM. The apolipoprotein A5 locus is a strong determinant of plasma triglyceride concentrations across ethnic groups in Singapore. J Lipid Res. 2003; 44: 2365–2373.

Ribalta J, Figuera L, Fernandez-Ballart J, Vilella E, Castro Cabezas M, Masana L, Joven J. Newly identified apolipoprotein AV gene predisposes to high plasma triglycerides in familial combined hyperlipidemia. Clin Chem. 2002; 48: 1597–1600.

C polymorphism in the apolipoprotein A5 gene is associated with postprandial hypertriacylglycerolemia;elevated small, dense LDL concentrations;and oxidative stress in nonobese Korean men. Am J Clin Nutr. 2004; 80 (4): 832–840.

Martin S, Nicaud V, Humphries SE, Talmud PJ. EARS group. Contribution of APOA5 gene variants to plasma triglyceride determination and to the response to both fat and glucose tolerance challenges. Biochim Biophys Acta. 2003; 1637: 217–225.

Hsu LA, Ko YL, Chang CJ, Hu CF, Wu S, Teng MS, Wang CL, Ho WJ, Ko YS, Hsu TS, Lee YS. Genetic variations of apolipoprotein A5 gene is associated with the risk of coronary artery disease among Chinese in Taiwan. Atherosclerosis. 2006; 185: 143–149.

C: association with plasma lipids and type 2 diabetes mellitus with coronary heart disease in Chinese. Clin Chem Lab Med. 2005; 43 (6): 607–612.

Lai CQ, Corella D, Demissie S, Cupples LA, Adiconis X, Zhu Y, Parnell LD, Tucker KL, Ordovas JM. Dietary intake of n-6 fatty acids modulates effect of apolipoprotein A5 gene on plasma fasting triglycerides, remnant lipoprotein concentrations, and lipoprotein particle size: the Framingham Heart Study. Circulation. 2006; 113: 2062–2070.

Vu-Dac N, Gervois P, Jakel H, Nowak M, Bauge' E, Dehondt H, Staels B, Pennacchio LA, Rubin EM, Fruchart–Najib J, Fruchart JC. Apolipoprotein A5, a crucial determinant of plasma triglyceride levels, is highly responsive to peroxisome proliferator-activated receptor activators. J Biol Chem. 2003; 278: 17982–17985.

Otvos JD. Measurement of lipoprotein subclass profiles by nuclear magnetic resonance spectroscopy. Clin Lab. 48 (3–4): 171–180, 2002.

Tsai MY, Georgopoulos A, Otvos JD, Ordovas JM, Hanson NQ, Peacock JM, Arnett DK. Comparison of ultracentrifugation and nuclear magnetic resonance spectroscopy in the quantification of triglyceride-rich lipoproteins after an oral fat load. Clin Chem. 2004; 50 (7): 1201–1204.

Livak KJ. 1999. Allelic discrimination using fluorogenic probes and the 5' nuclease assay. Genet Anal. 14; 143–9.

Littell RC, Milliken GA, Stroup WW, and Wolfinger RD 1996. SAS System for Mixed Models. Cary, NC, SAS Institute, Inc.

Singer JD. Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. J Educational and Behavioral Stat. 1998; 24: 323–355.

Merkel M, Loeffler B, Kluger M, Fabig N, Geppert G, Pennacchio LA, Laatsch A, Heeren J. Apolipoprotein AV accelerates plasma hydrolysis of triglyceride-rich lipoproteins by interaction with proteoglycan-bound lipoprotein lipase. J Biol Chem. 2005; 280 (22): 21553–60.

Vaessen SF, Schaap FG, Kuivenhoven JA, Groen AK, Hutten BA, Boekholdt SM, Hattori H, Sandhu MS, Bingham SA, Luben R, Palmen JA, Wareham NJ, Humphries SE, Kastelein JJ, Talmud PJ, Khaw KT. Apolipoprotein A-V, triglycerides and risk of coronary artery disease: the prospective Epic-Norfolk Population Study. J Lipid Res. 2006; 47 (9): 2064–70.

Merkel M, Heeren J. Give me A5 for lipoprotein hydrolysis! J Clin Invest. 2005; 115 (10): 2694–6.

Otvos JD, Collins D, Freedman DS, Shalaurova I, Schaefer EJ, McNamara JR, Bloomfield HE, Robins SJ. Low-density lipoprotein and high-density lipoprotein particle subclasses predict coronary events and are favorably changed by gemfibrozil therapy in the Veterans Affairs High-Density Lipoprotein Intervention Trial. Circulation. 2006; 113 (12): 1556–63.

Li WW, Dammerman MM, Smith JD, Metzger S, Breslow JL, Leff T. Common genetic variation in the promoter of the human apo CIII gene abolishes regulation by insulin and may contribute to hypertriglyceridemia. J Clin Invest. 1995; 96: 2601–2605.

作者单位：Nutrition and Genomics Laboratory (C.-Q.L., X.A., L.D.P., J.M.O.), JM-USDA Human Nutrition Research Center on Aging at Tufts University, Boston, Mass; the Department of Epidemiology (D.K.A.), University of Alabama at Birmingham; the School of Medicine (D.C.), University of Valencia, Spain; the Depar