SEARCH

SEARCH BY CITATION

Keywords:

  • association study;
  • cholesterol metabolism;
  • coronary heart disease;
  • peroxisome proliferator activated receptor delta;
  • polymorphism

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Objectives.  Peroxisome proliferator activated receptor delta (PPARD) is a transcription factor implicated in the regulation of genes involved in cholesterol metabolism. We recently discovered a common polymorphism in the 5′-untranslated region (5′-UTR) of the human PPARD, +294T/C, that is associated with an increased plasma low-density lipoprotein cholesterol (LDL-C) concentration in healthy subjects. Whether the +294C allele is associated with LDL-C elevation independently of the background lipoprotein phenotype and whether it confers increased risk of coronary heart disease (CHD) is unknown. Against this background, we investigated the relationships between the PPARD polymorphism and plasma lipoprotein concentrations and the risk for contracting CHD in the West of Scotland Coronary Prevention Study (WOSCOPS).

Design.  A nested case–control study of participants in a randomized double-blind placebo-controlled trial of pravastatin in mildly-to-moderately hypercholesterolaemic men.

Subjects.  A total of 580 cases of incident CHD and 1160 individuals who remained free of CHD (controls).

Main outcome measures.  Plasma lipoprotein con-centrations and risk of CHD according to PPARD genotype.

Results.  Individuals carrying the rare PPARD +294C allele had a significantly lower high-density lipoprotein cholesterol (HDL-C) concentration than subjects homozygous for the common T-allele. Homozygous carriers of the C-allele also showed a tendency towards higher risk of CHD compared with homozygous carriers of the T-allele. In addition, a gene–gene interaction involving the PPARD polymorphism and the PPAR alpha L162V polymorphism may influence the plasma LDL-C concentration.

Conclusions.  PPARD plays a role in cholesterol metabolism in man.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Epidemiological studies have provided strong evidence that an elevated plasma concentration of low-density lipoprotein cholesterol (LDL-C) is associated with an increased risk of developing coronary heart disease (CHD) [1]. Other major CHD risk factors include age, male gender, arterial hypertension, diabetes, smoking, a family history of premature CHD and low plasma high-density lipoprotein cholesterol (HDL-C) level [2]. Genetic factors play a significant role in the determination of plasma LDL-C and HDL-C concentrations [3]. Other factors that are clearly associated with dyslipoproteinaemia, particularly the low plasma HDL-C trait, are abdominal obesity [4], sedentary lifestyle and cigarette smoking. Data obtained from family and twin studies indicate that genetic factors play a major role in the susceptibility to CHD, and it is thought that variability in multiple genes acts in concert to influence the susceptibility to atherosclerosis [5].

The peroxisome proliferator activated receptors (PPARs) are potent transcription factors and dietary lipid sensors that regulate fatty acid and carbohydrate metabolism. Three subtypes, designated PPAR alpha (PPARA), PPAR gamma (PPARG) and PPAR delta (PPARD), are found in species ranging from Xenopus to humans [6]. They are encoded by separate genes and characterized by distinct tissue and developmental distribution patterns. Animal studies suggest that PPARD is involved in cholesterol metabolism, as treatment with PPARD agonists has been shown to increase plasma HDL-C levels in db/db mice and in obese rhesus monkeys [7, 8]. Furthermore, the potent selective PPARD agonist, GW501516, proved to decrease fasting plasma triglycerides and the fraction of small and dense LDL particles in obese rhesus monkeys [8].

We have recently discovered a common polymorphism in the 5′-untranslated region (5′-UTR) in PPARD at position +294. The effect of the +294T/C polymorphism was evaluated by association studies in two independent cohorts of a total of 825 healthy middle-aged men and by in vitro analyses such as electrophoretic mobility shift assay (EMSA) and transient transfection studies to study aspects of transcriptional regulation. The rare C-allele was shown to be associated with a significant increase in plasma LDL-C concentration, and the transcription factor Sp1 bound specifically to the C-allele which differentially regulated the transcriptional activity in vitro in human monocytic cells and in Sp1 deficient Drosophila cells [9]. In addition, evidence was obtained for an interaction with the PPARA L162V polymorphism that influences several plasma lipid parameters.

Against this background, we investigated the relationships of the PPARD +294T/C and the PPARA L162V polymorphisms to plasma lipoprotein concentrations, inflammatory markers and risk of contracting CHD in the West of Scotland Coronary Prevention Study (WOSCOPS). WOSCOPS was a prospective double-blind placebo-controlled study that evaluated the value of pravastatin in the prevention of coronary events and enrolled originally a total of 6595 hypercholesterolaemic men. Design features and methods of the WOSCOPS have been presented elsewhere [10–12]. The present investigation is based on a nested case–control study of a subset of individuals; the 580 subjects who contracted an event in the course of the study and subjects (controls) who showed no sign of CHD during the study (n = 1160).

Subjects

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Subjects were a total of 580 cases of incident CHD in WOSCOPS and a total of 1160 individuals who remained free of CHD (controls) who were examined in a nested case–control design. Briefly, WOSCOPS randomized 6595 unrelated caucasian men aged 45–64 years with LDL-C levels between 4.5 and 6.0 mmol L−1, to receive pravastatin 40 mg or placebo daily. Subjects had normal renal and hepatic function and no history of myocardial infarction or organ transplantation. The average follow-up period was 4.9 years. The primary end-point of the study was the occurrence of nonfatal myocardial infarction or death from CHD as a first event. The control subjects selected at random from the cohort were matched on the basis of age and smoking status. The study was approved by the ethics committees of the University of Glasgow, UK, all participating health boards and the Karolinska Hospital, Stockholm, Sweden.

Measurements

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

All major risk factors were assessed during recruitment [10]. Plasma lipids and lipoprotein concentrations (plasma triglycerides, plasma cholesterol, and VLDL, LDL and HDL cholesterol) were measured twice during screening according to the Lipid Research Clinics protocol [13]. Subjects were included in the study if they had plasma LDL-C ≥4.0 mmol L−1 on both, or ≥4.5 mmol L−1, on one occasion. If LDL-C exceeded 6.0 mmol L−1 at both screening visits, patients were excluded.

Blood counts were performed on Coulter STKR or S +1 automated cell counters. High-shear whole-blood viscosity was determined according to the equation of Whittington and Harkness [14] from data previously collected from West of Scotland men aged 45–64 years in the Scottish Heart Health Study/Glasgow MONICA Study [15]. C-reactive protein (CRP) and lipoprotein-associated phospholipase A2 (PLA2) were measured in aliquots of plasma collected at the third screening visit. A high sensitivity, two-time enzyme-linked immunoassay was developed with use of a peroxidase-conjugated rabbit antihuman CRP antibody (DK2600; Dako, Glostrup, Denmark) and a polyclonal anti-CRP capture antibody. The assay was calibrated with a standard (CRM470 – CAP/IFCC, lot 91/0619; Behringwerke, Marburg, Germany). The lower limit of the working range of the assay was 0.1 mg L−1. Values obtained in this study ranged from 0.1 to 45.2 mg L−1. The intra- and inter-assay coefficients of variation were 1.9 and 6.2%, respectively. Lipoprotein-associated PLA2 mass was measured with an enzyme-linked immunoassay as described [16] using a monoclonal capture antibody against lipoprotein-associated PLA2 and a second monoclonal antibody labelled with biotin and a streptavidin-alkaline phosphatase conjugate to identify the enzyme. The range of detection was 0.5–6.0 mg L−1, and the intra- and inter-assay coefficients of variation were 4.5 and 8.3%, respectively. There was no cross-reactivity with other A2 phospholipases. Fibrinogen was assayed by heat-precipitation nephelometry [17].

Genotyping

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Genotyping for the PPARD +294T/C polymorphism was performed using PCR amplification of genomic DNA and subsequent digestion by the restriction enzyme, Bsl I (New England Biolabs, Hertfordshire, UK), as described [9]. The analysis of the L162V PPARA polymorphism was performed by PCR amplification followed by real-time sequencing using the Pyrosequencing equipment (Pyrosequencing AB, Uppsala, Sweden), as described [9].

Due to missing samples [n = 117 (PPARD) and n = 121 (PPARA)] and difficulties in amplifying by PCR [n = 4 (PPARD) and n = 2 (PPARA)], genotype results were obtained from 501 cases and 1118 and 1116 control subjects for the PPARD and PPARA polymorphisms, respectively.

Statistical analyses

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Allele frequencies were estimated by gene counting. A chi-square test was used to compare the observed numbers of PPARD genotypes with those expected for a population in Hardy–Weinberg equilibrium. Categorical variables were summarized by counts and percentages and the continuous variables were summarized by mean and standard deviations. The sums were calculated separately for cases and controls according to PPARD genotype. Both triglycerides and CRP were log-transformed before statistical testing, because of skewed distributions.

The association between the PPARD +294TT genotype and cardiovascular risk was analysed by conditional logistic regression, which takes into account the matching criteria. In addition to the univariate model, three adjusted models were created. Model (1) adjusted for traditional risk factors such as age, systolic blood pressure, plasma total triglyceride concentration, and plasma LDL-C and HDL-C concentrations. In addition to these, model (2) also adjusted for inflammatory markers such as CRP, fibrinogen and white cell count. A third model added in PLA2 as an additional factor for adjustment. The results of the logistic models are presented in the form of odds ratios (with 95% confidence limits) for the genotype groups using the TT genotype as the referent level. Interaction tests between the PPARD genotype and the effects of smoking, alcohol, and treatment with pravastatin and the PPARA genotype were carried out using likelihood ratio tests. The risk factors were compared amongst the genotypes using anova.

PPARD genotype distributions and risk of CHD

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

The frequency of the rare C-allele was 0.21 amongst the cases and 0.18 in the control group (P = 0.088) (Table 1). The distributions of genotypes were similar between cases and control subjects and corresponded to those expected for populations in Hardy–Weinberg equilibrium. An imbalance in missing samples between the two groups adversely affected the power to detect a statistically significant difference between the groups. Almost twice as many samples were missing from the case group compared with the control group (n = 79 and n = 42, respectively). Of note, the difference in missing samples between the case and control groups was not due to selection but merely to insufficient amount of DNA. Regardless of pravastatin treatment the genotype distribution is similar both in cases and controls. Accordingly, cases and controls were analysed irrespective of whether pravastatin or placebo treatment had been instituted.

Table 1.  Genotype distributions in cases and controls
 Cases (%)Controls (%)
Number of subjects5011118
PPARD +294TT317 (63.3)747 (66.8)
PPARD +294TC159 (31.7)333 (29.8)
PPARD +294CC25 (5.0)38 (3.4)

To examine whether PPARD genotype influenced the risk of CHD homozygote carriers of the T and C-alleles were compared. Homozygotes for the rare C-allele showed a tendency towards higher risk of CHD compared with homozygous carriers of the common T-allele, with an odds ratio (OR) of 1.52 (95% CI 0.91–2.57). Subjects heterozygous for the polymorphism showed an OR of 1.13 (95% CI 0.90–1.43). Control for the confounding effects of clinical and lipoprotein risk factors [model 1, TT vs. CC; OR of 1.56 (95% CI 0.91–2.67)] and inflammatory markers [models 2 and 3, TT vs. CC; ORs of 1.54 (95% CI 0.86–2.73) and 1.48 (95% CI 0.83–2.64), respectively], did not influence the relationship of PPARD genotype to CHD. The risk of CHD in subjects carrying either one or two copies of the C-allele was OR 1.18 (95% CI 0.94–1.47).

Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Plasma concentrations of lipids and major lipoproteins were assessed according to PPARD +294T/C genotype. Due to the low number of homozygous for the C-allele we compared carriers of one or two copies of the rare C-allele with individual homozygotes for the common T-allele. In the control group, carriers of the rare C-allele showed significantly lower HDL-C concentration than subjects homozygous for the common T-allele [mean (SD): CC 1.11 (0.25); TC 1.12 (0.25); TT 1.15 (0.25) mmol L−1; P = 0.049] (Table 2). No associations were found between the +294T/C polymorphism and plasma concentrations of other lipids or lipoproteins. Furthermore, no genotype associations were found amongst the cases with plasma lipid or lipoprotein concentrations (data not shown). In neither controls nor cases were there associations between PPARD genotype and body mass index (BMI) or inflammatory markers such as PLA2, CRP, fibrinogen and white cell count (Table 2). Examination of the relationships of PPARD genotype to plasma lipid and lipoprotein concentrations in subjects below or above the median value for BMI (25.3) revealed no differences in associations.

Table 2.  Associations between the +294T/C polymorphism in exon 4 of PPARD and the PPARA L162V polymorphism with body mass index (BMI), plasma and major lipoprotein lipids, and inflammatory markers and interactive effects between the PPAR genotypes in the control group of WOSCOPS
 PPARD +294TTPPARD +294TCPPARD +294CCPPARD TT vs. TC/CC (P-value)PPARA L162V (P-value)Interaction (P-value)
  1. VLDL, very-low density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; PLA2, lipoprotein-associated phospholipase A2; CRP, C-reactive protein; WCC, white cell count.

  2. *The P-values derive from data sets that have been log-transformed.

  3. Values are given as mean ± SD.

Number of subjects74733338   
BMI (kg m−2)25.61 ± 3.2525.73 ± 3.0725.46 ± 3.630.6580.9160.341
Cholesterol (mmol L−1)
 Plasma7.02 ± 0.577.05 ± 0.596.93 ± 0.450.5900.1660.331
 VLDL0.86 ± 0.380.89 ± 0.400.84 ± 0.290.3300.4210.480
 LDL4.94 ± 0.444.98 ± 0.464.92 ± 0.330.2390.1910.029
 HDL1.15 ± 0.251.12 ± 0.251.11 ± 0.250.0490.6670.480
Triglycerides (mmol L−1)1.82 ± 0.771.88 ± 0.771.82 ± 0.640.179*0.445*0.917*
PLA2 (mg L−1)2.27 ± 0.572.26 ± 0.552.36 ± 0.650.9720.5340.467
CRP (mg L−1)3.31 ± 4.253.47 ± 5.162.18 ± 1.980.767*0.552*0.589*
Fibrinogen (g L−1)4.35 ± 0.854.38 ± 0.894.24 ± 0.690.8500.3940.702
WCC (109 L−1)6.76 ± 1.856.79 ± 1.946.33 ± 1.370.8810.0830.874

Evaluation of gene–treatment and gene–environment interactions

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

WOSCOPS allowed us to evaluate whether PPARD genotype might influence the response to pravastatin treatment. Overall, there was no significant pravastatin treatment effect according to genotype group on lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers in neither cases or controls.

Associations between PPARD genotype and risk of CHD stratified on alcohol use and smoking habits were also evaluated. There was a trend towards an association between PPARD genotype and CHD risk when subjects were analysed according to alcohol consumption; the OR for +294TT compared with +294CC was higher for nondrinkers (OR = 2.94, 95% CI 0.89–9.65, interaction test; P = 0.078). Smoking habits, on the other hand, did not influence the relationship between PPARD genotype and risk of CHD.

Evaluation of gene–gene interaction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Since statistical evidence of an interaction between the PPARD +294T/C and the PPARA L162V polymorphisms was noted for several lipid parameters amongst healthy middle-aged men, interactive effects between the PPAR genotypes were investigated in WOSCOPS. The frequency of the rare PPARA 162V allele was 0.077 amongst the cases and 0.068 in the control group and did not influence the risk of CHD (data not shown). The pravastatin treatment effect did not differ according to PPARA L162V genotype group in either cases or controls. Accordingly, cases and controls were analysed irrespective of pravastatin or placebo treatment. Due to the low number of homozygotes for the rare PPARA 162V allele we compared carriers of the rare V allele with individuals homozygous for the common L allele. From a statistical point of view, interaction between the two PPAR genotypes had a significant effect on plasma LDL-C (P = 0.029, Table 3). No interactions were detected between PPAR genotypes in relation to other plasma lipid or lipoprotein concentrations, BMI or inflammatory markers. Of note, no associations were observed between the PPARA polymorphism alone and plasma concentrations of lipids or lipoproteins, BMI and inflammatory markers (Table 2).

Table 3.  Interaction between the PPARD +294T/C and the PPARA L162V polymorphism on plasma LDL-C concentration in the control group of WOSCOPS
PPARA L162VPPARD +294T/CLDL-C (mmol L−1)n
  1. LDL-C, plasma low-density lipoprotein concentration; n, number of subjects. Values are given as mean ± SD.

LLTT4.96 ± 0.44647
TC4.97 ± 0.45297
CC4.89 ± 0.3529
LV/VVTT4.85 ± 0.4098
TC5.05 ± 0.5336
CC5.02 ± 0.239

Although, an interactive effect was indicated between the PPAR genotypes on plasma LDL-C concentration, this effect did not influence the relationship of PPAR genotypes to CHD. Thus, inclusion of the PPARA L162V polymorphism into the model did not modulate the risk of CHD [OR of 1.48 (95% CI 0.45–4.85)].

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

To address the question of whether the common PPARD +294T/C polymorphism influences lipid and lipoprotein phenotypes, inflammatory markers, pravastatin treatment effect and risk of CHD we chose to investigate a group of individuals already at high risk of CHD from WOSCOPS. The results provide some further evidence that PPARD plays a role in human cholesterol metabolism.

In this nested case–control study, carriers of the rare C-allele of the +294T/C polymorphism had significantly lower plasma HDL-C concentrations compared with homozygous T-allele carriers, suggesting a role for this gene in the regulation of HDL-C levels in hypercholesterolaemia. In a separate recent study, we have observed an association between the rare C-allele and higher plasma LDL-C concentrations in normolipidaemic healthy middle-aged men [9]. The lack of an association between PPARD genotype and LDL-C levels in the present report is likely to be due to the restricted LDL-C inclusion criterion used in WOSCOPS (LDL-C of 4.5–6.0 mmol L−1). In contrast, no restriction regarding plasma HDL-C concentrations was applied [11], and the values seen in the recruited men reflected the distribution observed in the general population (J. Skogsberg and E. Ehrenborg, unpublished observation). Given the multiple roles that PPARs play in lipid metabolism, it is perhaps not surprising that elevated PPARD expression (presumed to be present in PPARD +294C carriers) results in a different lipoprotein phenotype in hypercholesterolaemic compared with normocholesterolaemic individuals.

The role of PPARD in lipid metabolism is not yet fully understood, but several animal studies have provided evidence for an involvement of PPARD in cholesterol metabolism. Treatment with potent PPARD agonists results in increased HDL-C levels in both db/db mice [7] and obese rhesus monkeys [8]. Since the +294C allele is associated with an increase in transcriptional activity in vitro in transient transfection studies, one would anticipate an association between the C-allele and higher HDL-C concentrations in WOSCOPS. The seemingly paradoxical finding of a lower HDL-C concentration in carriers of the +294C allele could be explained by several obvious differences between the present study and the previously reported animal studies. First, it is important to keep in mind that our studies show the effects of one single polymorphic site on lipid and lipoprotein phenotypes in clinical cohorts in a genotype–phenotype association study setting whilst the animal studies evaluate the treatment effects of potent PPARD agonists. It is therefore not surprising that we found relatively small effects. However, association studies in well-characterized populations are powerful tools to elucidate physiological roles in vivo of particular genes. Secondly, the in vitro studies performed in our previous paper [9] were conducted in a human monocytic cell line. Whereas monocytes play a central role in the atherosclerotic process, they are much less important in plasma cholesterol metabolism. This might account for the seemingly discrepant findings of the association and in vitro studies. Studies in a human hepatic cell line would have been more appropriate in the context of cholesterol metabolism as PPARD effects might differ between cell lines. Thirdly, the BMI of the participants in WOSCOPS was within the normal range; only 95 of 1118 individuals had a BMI above 30. Of note, both the db/db mice and the rhesus monkeys were obese and insulin-resistant, which influences the energy homeostasis and may result in different PPARD relationships to lipid and lipoprotein parameters. In this study, no interactions between PPARD genotype and BMI on lipid parameters were observed. Taken together, it is in fact notable that both our studies and the animal studies demonstrate involvement of PPARD in cholesterol metabolism.

The PPARA L162V polymorphism has also been implicated in cholesterol metabolism [18–20]. In this study of hypercholesterolaemic men, the statistical analysis indicated that an interaction between the PPARD +294T/C and the PPARA L162V polymorphisms may influence the LDL-C concentration, although the individual PPAR genotypes were unassociated with this parameter. In our recent study of healthy middle-aged men we observed similar indications of interactions between the PPAR genotypes on LDL-C, HDL-C and triglyceride concentrations [9]. The lack of interactive effects on other plasma lipid or lipoprotein concentrations in WOSCOPS could be explained by the fact that the men in this study were selected on the basis of a rather high LDL-C concentration. The effect of the PPARA polymorphism has been shown to depend on the metabolic status [20, 21]. In addition, the hypercholesterolaemic men in WOSCOPS have a high risk factor burden, including higher levels of inflammatory markers and increased plasma triglyceride concentration, which also alter the metabolic condition. Furthermore, different ethnicity might also contribute to the discrepancy between the two studies. It is notable that both this and our previous study in healthy middle-aged men [9] lacked sufficient power to assess the biological significance of the statistical interactions observed.

Statins are inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in cholesterol synthesis. Statin treatment has been shown to influence PPAR activity in vitro. PPARG is activated by the cholesterol depletion induced by statins, which triggers the cleavage of cholesterol-sensitive transcription factors called SREBPs, resulting in an increase of transcriptional activity [22] and production of endogenous ligands for the receptor [23]. PPARA, on the other hand, is activated by statins through a completely different pathway involving the Rho family of proteins [24]. In contrast, it is unknown whether PPARD is influenced by statins in vivo. WOSCOPS allowed us to explore whether gene–drug interactions exist in relation to the PPARD +294T/C polymorphism and statin treatment amongst moderately hypercholesterolaemic men. The +294T/C polymorphism is located in a region of potential interest for gene expression, as the transcription factor Sp1 binds to this site [9]. However, we could not detect any effect of this polymorphism on the response to pravastatin. Nevertheless one cannot exclude the possibility that statins could alter PPARD mRNA and protein levels as well as the activation of the receptor.

PPARs play modulatory roles in the inflammatory process [25]. That PPARs are implicated in inflammation was first indicated by the prolonged inflammatory response in PPAR alpha null mice [26]. Furthermore, PPAR activators have been shown to inhibit cytokines and matrix metalloproteinases by repression of NF-κB, AP-1 and STAT-1 signalling pathways [27–28]. In WOSCOPS markers of inflammation, such as PLA2, CRP, fibrinogen and white-cell count, were associated with the risk of CHD, and PLA2 proved to be an independent predictor of coronary events [29]. Whether PPARD is implicated in inflammatory processes remains unclear, but data from Ppard−/− mice show that the response to nonsteroid anti-inflammatory drug treatment is dependent on PPARD [30]. However, the data obtained from this study did not provide evidence of an influence of PPARD expression on inflammation based on the absense of any associations between inflammatory markers and the PPARD +294T/C polymorphism.

In conclusion, this study suggests that PPARD is implicated in cholesterol metabolism in humans based on the association between the PPARD +294T/C polymorphism and the plasma HDL concentration, a well-established risk factor for coronary events. The PPARD +294T/C polymorphism also tended to modulate CHD risk. In addition, a gene–gene interaction involving PPARD and PPARA was indicated to influence the plasma LDL-C concentration. One might speculate that other genetic variants or specific activators of PPARD influence the development of atherosclerosis.

Acknowledgements

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References

Members of the West of Scotland Coronary Prevention Study Group are Stuart Cobbe, Ian Forde, P Macfarlane, AR Lorimer, Chris Packard, Jim Shepherd for the University of Glasgow & C Isles for Dumfries & Galloway Royal Infirmary, Scotland.

This work was supported by the Swedish Medical Research Council, the Swedish Heart–Lung Foundation, Swedish Diabetes Foundation, the National Board of Health and Welfare and the following foundations: Foundation for Geriatric Research, Åke Wiberg, Sigurd and Elsa Golje Memorial, Fredrik and Ingrid Thuring, Prof. Nanna Svartz, the Old Servants and the Karolinska Institute. J. Skogsberg is supported by doctoral research fellowship from the National Network for Cardiovascular Research.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Subjects
  6. Measurements
  7. Genotyping
  8. Statistical analyses
  9. Results
  10. PPARD genotype distributions and risk of CHD
  11. Association of PPARD genotype with lipid and lipoprotein concentrations, anthropometric measurements and inflammatory markers
  12. Evaluation of gene–treatment and gene–environment interactions
  13. Evaluation of gene–gene interaction
  14. Discussion
  15. Conflict of interest statement
  16. Acknowledgements
  17. References
  • 1
    Wilson PW, Garrison RJ, Castelli WP, Feinleib M, McNamara PM, Kannel WB. Prevalence of coronary heart disease in the Framingham Offspring Study: role of lipoprotein cholesterols. Am J Cardiol 1980; 46: 64954.
  • 2
    Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 183747.
  • 3
    Hamsten A, Iselius L, Dahlen G, de Faire U. Genetic and cultural inheritance of serum lipids, low and high density lipoprotein cholesterol and serum apolipoproteins A-I, A-II and B. Atherosclerosis 1986; 60: 199208.
  • 4
    Despres JP. Dyslipidaemia and obesity. Baillieres Clin Endocrinol Metab 1994; 8: 62960.
  • 5
    Marenberg ME, Risch N, Berkman LF, Floderus B, de Faire U. Genetic susceptibility to death from coronary heart disease in a study of twins. N Engl J Med 1994; 330: 104146.
  • 6
    Willson TM, Brown PJ, Sternbach DD, Henke BR. The PPARs: from orphan receptors to drug discovery. J Med Chem 2000; 43: 52750.
  • 7
    Leibowitz MD, Fievet C, Hennuyer N et al. Activation of PPARdelta alters lipid metabolism in db/db mice. FEBS Lett 2000; 473: 33336.
  • 8
    Oliver WR, Shenk JL, Snaith MR et al. A selective peroxisome proliferator-activated receptor delta agonist promotes reverse cholesterol transport. Proc Natl Acad Sci USA 2001; 98: 530611.
  • 9
    Skogsberg J, Kannisto K, Cassel TN, Hamsten A, Eriksson P, Ehrenborg E. Evidence that peroxisome proliferator-activated receptor delta influences cholesterol metabolism in men. Arterioscler Thromb Vasc Biol 2003; 23: 63743.
  • 10
    The West of Scotland Coronary Prevention Study Group. Screening experience and baseline characteristics in the West of Scotland Coronary Prevention Study. The WOSCOPS Study Group. West of Scotland Coronary Prevention Study. Am J Cardiol 1995; 76: 48591.
  • 11
    The West of Scotland Coronary Prevention Study Group. A coronary primary prevention study of Scottish men aged 45–64 years: trial design. The West of Scotland Coronary Prevention Study Group. J Clin Epidemiol 1992; 45: 84960.
  • 12
    Shepherd J, Cobbe SM, Ford I et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995; 333: 130107.
  • 13
    Lipid Research Clinics Program. Lipid and Lipoprotein Analysis: Manual of Laboratory Operations (Publication NIH/75–628). Washington DC: Department of Health, Education and Welfare, 1982.
  • 14
    Whittington RB, Harkness J. Whole-blood viscosity, as determined by plasma viscosity, haematocrit, and shear. Biorheology 1982; 19: 17584.
  • 15
    Lowe G, Smith W, Tunstall-Pedoe H. Cardiovascular risk and haemorheology: results from the Scottish Heart Health Study and the MONICA-project, Glasgow. Clin Hemorheol 1988; 8: 51824.
  • 16
    Caslake MJ, Packard CJ, Suckling KE, Holmes SD, Chamberlain P, Macphee CH. Lipoprotein-associated phospholipase A(2), platelet-activating factor acetylhydrolase: a potential new risk factor for coronary artery disease. Atherosclerosis 2000; 150: 41319.
  • 17
    Yarnell JW, Baker IA, Sweetnam PM et al. Fibrinogen, viscosity, and white blood cell count are major risk factors for ischemic heart disease. The Caerphilly and Speedwell collaborative heart disease studies. Circulation 1991; 83: 83644.
  • 18
    Tai ES, Demissie S, Cupples LA et al. Association between the PPARA L162V polymorphism and plasma lipid levels: the Framingham Offspring Study. Arterioscler Thromb Vasc Biol 2002; 22: 80510.
  • 19
    Vohl MC, Lepage P, Gaudet D et al. Molecular scanning of the human PPARa gene: association of the L162v mutation with hyperapobetalipoproteinemia. J Lipid Res 2000; 41: 94552.
  • 20
    Flavell DM, Pineda Torra I, Jamshidi Y et al. Variation in the PPARalpha gene is associated with altered function in vitro and plasma lipid concentrations in Type II diabetic subjects. Diabetologia 2000; 43: 67380.
  • 21
    Jamshidi Y, Flavell DM, Hawe E, MacCallum PK, Meade TW, Humphries SE. Genetic determinants of the response to bezafibrate treatment in the lower extremity arterial disease event reduction (LEADER) trial. Atherosclerosis 2002; 163: 18392.
  • 22
    Fajas L, Schoonjans K, Gelman L et al. Regulation of peroxisome proliferator-activated receptor gamma expression by adipocyte differentiation and determination factor 1/sterol regulatory element binding protein 1: implications for adipocyte differentiation and metabolism. Mol Cell Biol 1999; 19: 5495503.
  • 23
    Kim JB, Wright HM, Wright M, Spiegelman BM. ADD1/SREBP1 activates PPARgamma through the production of endogenous ligand. Proc Natl Acad Sci USA 1998; 95: 433337.
  • 24
    Martin G, Duez H, Blanquart C et al. Statin-induced inhibition of the Rho-signaling pathway activates PPARalpha and induces HDL apoA-I. J Clin Invest 2001; 107: 142332.
  • 25
    Chinetti G, Fruchart JC, Staels B. Peroxisome proliferator-activated receptors (PPARs): nuclear receptors at the crossroads between lipid metabolism and inflammation. Inflamm Res 2000; 49: 497505.
  • 26
    Devchand PR, Keller H, Peters JM, Vazquez M, Gonzalez FJ, Wahli W. The PPARalpha-leukotriene B4 pathway to inflammation control [see comments]. Nature 1996; 384: 3943.
  • 27
    Ricote M, Li AC, Willson TM, Kelly CJ, Glass CK. The peroxisome proliferator-activated receptor-gamma is a negative regulator of macrophage activation. Nature 1998; 391: 7982.
  • 28
    Straus DS, Pascual G, Li M et al. 15-deoxy-delta 12,14-prostaglandin J2 inhibits multiple steps in the NF-kappa B signaling pathway. Proc Natl Acad Sci USA 2000; 97: 484449.
  • 29
    Packard CJ, O'Reilly DS, Caslake MJ et al. Lipoprotein-associated phospholipase A2 as an independent predictor of coronary heart disease. West of Scotland Coronary Prevention Study Group. N Engl J Med 2000; 343: 114855.
  • 30
    Peters JM, Lee SS, Li W et al. Growth, adipose, brain, and skin alterations resulting from targeted disruption of the mouse peroxisome proliferator-activated receptor beta(delta). Mol Cell Biol 2000; 20: 511928.