Paraoxonase and coronary heart disease

https://doi.org/10.1016/S1567-5688(02)00046-6Get rights and content

Abstract

The antioxidant activity of HDL is largely due to the paraoxonase (PON1) located on it. Experiments with transgenic PON1 knock-out mice indicate the potential for PON1 to protect against atherogenesis. This effect of HDL in decreasing LDL lipid peroxidation is maintained for longer than that of antioxidant vitamins and could thus be more protective. Several important advances in the field of PON research have occurred recently, not least the discovery that two other members of the PON gene family PON2 and PON3 may also have important antioxidant properties. Significant advances have been made in understanding the basic biochemical function of PON1 and the discovery of possible modulators of its activity. Decreased coronary heart disease (CHD) risk associated with polymorphisms of PON1 which are most active in lipid peroxide hydrolysis revealed by meta-analysis is likely to be an underestimate of the true contribution of PON1 to CHD because these polymorphisms explain only a small component of the variation in PON1 activity. However, it is a very important observation because genetic influences are not likely to be confounded by other factors linked with both CHD and diminished PON1 activity. PON1 is extensively researched and strategies will hopefully emerge to increase its activity and provide a more satisfactory test of the antioxidant hypothesis of atherosclerosis than antioxidant vitamins have done.

Introduction

Among the risk factors for coronary heart disease (CHD) identified by epidemiological studies, low plasma HDL concentration is one of the strongest [1]. For any given LDL concentration, the HDL-cholesterol concentration is inversely correlated with the risk of CHD and stroke [2], [3], [4].

The protective effect of HDL against the development of CHD appears to be complex. Much research in this field has centred on the lipid transport function of HDL particularly reverse cholesterol transport and whether any component of HDL aids the efflux of cholesterol from the artery wall. However, another area of interest has emerged in recent years: how HDL protects LDL and cell membranes against lipid peroxide-induced damage thus potentially impeding the initiation and progression of atheromatous lesions. The enzymes of HDL are critical for this and they and their involvement in atherosclerosis are the subject of this review.

The paraoxonase gene family has three known members, PON1, PON2 and PON3, located on the long arm of chromosome 7 between q21.3 and q22.1 in humans [5]. The genes share considerable structural similarity and appear to have arisen by gene duplication from a common evolutionary precursor. Within a given mammalian species, PON1, PON2 and PON3 share approximately 60% identity at the amino acid level and about 70% identity at the nucleotide level. However between mammalian species each of the three genes share 79–90% identity at the amino acid level and 81–91% identity on the nucleotide levels [5], [6]. Codon 106 (lysine) present in PON1 [5], [6] is missing in all PON2 and PON3 cDNA's sequenced to date. PON1 has been studied for many years following the discovery that the capacity of human serum to hydrolyse xenobiotics was due to its presence. More recently attempts have been made to identify its physiological substrates and to explore its relationship with atherosclerosis. PON2 and PON3 have only recently become the subject of investigation and are less well understood than PON1. For this reason only paraoxonase 1 (PON1) will be discussed in this review.

PON1 is an enzyme with a molecular mass of 43 KDa (354 amino acids) and in serum is exclusively located on HDL [7]. PON1 hydrolyses organophosphate substrates such as paraoxon. PON1 is most likely to explain the ability of HDL to metabolise lipid peroxides and to protect against their accumulation on LDL. Our initial report that this process was largely due to an enzyme or enzymes acting at a specific point in the lipid peroxidation cascade [8], [9] has been repeatedly confirmed [10], [11]. We found that purified PON1 was significantly more efficient than apo A1 or LCAT in preventing oxidation of LDL, although addition of the latter components did slightly enhance the PON1 effect [12]. PON1 can accelerate both the breakdown of phospholipid hydroperoxides [10], [11], [13] and platelet-activating factor (PAF) [14]. Potentially this generates lysolipids, aldehydes and ketones which can lead to atherogenic modification of LDL [15]. Indeed the PAFAH (PLA2) activity on LDL independently predicts CHD [16]. We have, however, recently provided evidence that PAFAH activity on HDL, unlike that on LDL, is due to PON1 and not to a separate enzyme (see later). Furthermore the location on HDL of LCAT which in humans generates substantial quantities of lysophosphatidylcholine (lyso-PC) each day suggests that this is safely accomplished on HDL as opposed to LDL. This may explain why an antioxidant enzyme system involving PON1 generating similar potentially toxic products has evolved on HDL. This is probably because our previous studies have shown a mechanism for the disposal of lyso-PC on HDL (Table 1). Furthermore, HDL, unlike LDL, is ubiquitously distributed throughout the tissue fluid where it is present at higher concentration than LDL or, as in the case of the CNS, the only lipoprotein present. It is likely therefore that its capacity to accept and to detoxify lipid peroxides provides general protection for all membranes and that it is the resemblance of LDL to a cell membrane that explains its capacity to protect LDL from oxidative modification [15]. It has been shown to protect erythrocyte membranes equally well [17] HDL from the PON1 knock-out mouse [18] and from non-mammalian species lacking PON1, such as birds [19] and HDL in the presence of PON1 inhibitors [11] loses its capacity to impede the accumulation of lipid peroxides under oxidising conditions.

PON1 has two coding region amino acid polymorphisms, one at position 55 (methionine/leucine; M/L) and another at position 192 (arginine/glutamine; R/Q) [7]. In addition to these coding region polymorphisms five others have been reported in the PON1 promoter region [20], [21], [22]; these polymorphisms are at −107/−108, −126, −160/−162, −824/−832 and −907/−909, using the system where the base immediately before the start codon is numbered as ‘−1’. Disagreement about nomenclature for four of the polymorphisms exists probably due to small variations in the sequences reported by different laboratories. Brophy et al. in cell culture experiments with a reporter gene found that −108, −162 and −909 polymorphisms had a functional effect on PON1 expression [23].

Of the PON 192 alloenzymes the R alloenzyme has proved to be more efficient at protecting LDL from oxidation. Numerous case-control studies have therefore been conducted to determine whether the PON1 192 R polymorphism is more closely associated with CHD than the Q polymorphism. These have all reported that either this is the case or that there was no association with either of the PON1 192 alleles [7]. A recent meta-analysis revealed a statistically significant increased likelihood of CHD associated with the PON1 192 R allele. Some studies suggest that the PON1 R allele may increase susceptibility to other established CHD risk factors, such as diabetes mellitus [24], cigarette-smoking [25] and age [26]. The PON 55 L alloenzyme is also more effective in vivo in protecting LDL against oxidation than the M alloenzyme. Few case-control studies of the 55 polymorphism have been done. Some have shown an association between the PON1 55 L allele and atherosclerosis [27], [28], but others have not [29], [30]. As yet there are no prospective investigations of CHD and PON1 polymorphisms. Also the association between CHD and PON1 genotype although largely confirmatory is not the only test of the hypothesis that PON1 protects against CHD. There is a substantial 40-fold, interindividual variation in PON1 activity, which is independent of either the 55 or 192 polymorphisms. This may be due to acquired factors acting either on the composition of the lipid environment of HDL, in which PON1 operates, or on the promoter region of the PON1 gene or in some manner as yet unidentified. When PON1 activity is measured directly in patients with CHD, it is about half that of disease-free controls [31], [32], [33]. This appears to be the case even within a few hours of the onset of cardiac ischaemic chest pain in myocardial infarction, suggesting that low serum PON1 activity may have preceded the event (Fig. 1) [32]. Low serum PON1 activity independent of genotype has been reported with several diseases, which are known to be associated with CHD, including clinical and experimental diabetes mellitus [24], [34], [35], [36], [37], hypercholesterolaemia [34] and renal failure [38].

We have shown that PON1 immunoreactivity is increasingly present in the arterial wall as atheroma advances [39]. There is no way at present of knowing whether this is part of a protective response, but a recent study has shown that PON1 has the ability ex vivo to hydrolyse lipid peroxides within human carotid and coronary atheromatous lesions [40].

New insights into the mechanism of the prevention of atherosclerosis by PON1 have been acquired recently. We have shown that purified human PON1 hydrolyses the pro-inflammatory mediator PAF and by using a selective inhibitor of PAF acetyl hydrolyase (LDL associated phospholipase A2), SB-222657, which does not inhibit PON1, that all the PAF hydrolysing activity of HDL was due to PON1 (Fig. 2) [14]. Ahmed and colleagues compared the oxidation products of native HDL, trypsinised HDL and HDL lipid suspensions and phosphatidylcholine apo AI proteoliposomes and found that apo AI increased the formation of phosphatidylcholine core aldehydes [41], [42]. Lysophosphatidylcholine was found in significant amounts only during oxidation of intact HDL, consistent with activation of a phospholipase A2 like activity. They concluded that PON1 has a phospholipase A2 like activity towards phosphatidylcholine core aldehydes.

Stable transfection of PON1 in Chinese hamster ovary (CHO) cells showed very little PON1 release in the absence of an appropriate acceptor, such as HDL [43]. Neither lipid-poor apo AI, apo AII nor LDL could substitute for HDL. The authors of the report hypothesised that PON1 release involved a docking process where HDL briefly links with the cell membrane and removes the peptide from the external membrane of the cell. This process may facilitate the stereochemical alignment of PON1.

Brushia et al. [44] used a baculovirus expression system to produce the large amounts of PON1 necessary for structure-function studies. These showed that glycosylation was not necessary for PON1 antioxidant activity, but was essential for its arylesterase activity. The same group transfected human PON1 into CHO cells and investigated apo AI cysteine substitutions. They found that for optimal PON1 activity, coassembly of the enzyme onto nascent HDL was required and that the N-terminal region of apo AI could be important in this assembly process [45].

A transgenic mouse over-expressing PON1 was developed by Oda et al. to determine whether PON1 could preserve HDL function during oxidative stress. They found that over-expression of PON1 inhibited lipid hydroperoxide formation on HDL and protected HDL integrity and function [46]. Van Lenten et al. have reported that oxidised phospholipids found in mildly oxidised LDL act acutely to decrease the expression of PON1 and increase that of apo J partly through the inflammatory cytokine, IL-6 [47]. IL-6 was critical to short-term, but not long-term, regulation of PON1 whereas IL-6 is not required for oxidised phospholipid regulation of MCP-1.

Unlike chain-breaking antioxidants, HDL prevents the accumulation of lipid peroxides on LDL [8] and in the vessel wall [40] for several hours, continuing to do so long after fat-soluble antioxidants have been exhausted (Table 2) and this activity appears to be due to PON1.

One of the appeals of PON1 is that the anti-atherogenic mechanism it offers could link LDL oxidation with atherogenesis despite the failure of antioxidant vitamin trials. If therefore, it turns out that antioxidant vitamins are themselves major determinants of PON1 activity, interest in PON1 could considerably diminish. However, there was a negative correlation between serum PON1 activity and the intake of vegetables, presumably containing large amounts of vitamins C and E [48]. None the less, Aviram and colleagues have previously shown in vitro that the antioxidant flavanoids, quercetin and glabridin, can protect PON1 in micellar solution (isolated from other HDL components) from loss of activity due to Cu2+-induced oxidation [49]. These, however, were in vitro observations and Arrol et al. were unable to influence PON1 activity even with pharmacological doses of vitamin E given to volunteers [50]. PON1 probably therefore does not require vitamin E for its activity except perhaps in quantities present basally in the participants in their study.

Other nutritional and environmental indicators of PON1 activity have been reported. High serum cholesterol [34] and insulin resistance [51] are, for example, associated with decreased PON1 activity. Studies of macronutrients in man have not so far been undertaken, but in rodents feeding monounsaturated fatty acids has been reported to lead to higher serum PON1 activity than saturated or highly polyunsaturated fatty acids [52]. Degraded cooking oil [53] and an atherogenic diet [54], [55] have also been reported to decrease PON1 activity in rabbits, mice and humans. Polyphenols (present in wine, tea and fruit juice) also increase PON1 activity in both man and mouse [56] as does moderate alcohol intake [57]. PON1 decreases in older people and with the menopause [34]. Acute exposure to organophosphates decreases PON1 activity [58], but it is not certain yet whether chronic occupational or environmental low level exposure to organophosphates or other toxins can influence PON1 activity.

Between populations there are also marked differences in PON1 activity, with populations of non-European origin having higher levels [59]. In part this is because of the higher activity 192 polymorphic alleles are prevalent in people of African and Asian origin. However, differences in nutrition and industrialisation could be important. Understandably most of the interest in pharmacological effects on PON1 activity has thus far been in effects of lipid-lowering drugs. Some of these studies [60], [61], [62], but not all [63], [64], [65], suggest an effect of statins and fibrates in raising PON1 activity. Seeking dietary or pharmacological interventions which will significantly increase PON1 activity could have an application to the prevention of arteriosclerosis and might make it possible to test the oxidant theory of atherosclerosis by an approach other than antioxidant vitamins.

Section snippets

Conclusion

The three known members of the PON gene family are all capable of impeding lipid peroxidation and could therefore act in an anti-atherosclerotic role. Most research has focussed on PON1. In case-control studies its serum activity is substantially decreased in CHD and, although explaining only a small part of the variation in PON1 activity, the PON1 192, polymorphism least able to prevent LDL oxidation in vitro is associated with a greater likelihood of CHD. Nutritional and pharmacological

Acknowledgements

The authors thank Ms Caroline Price for expert typing of the manuscript and all the friends and colleagues who have contributed to this work over the years.

References (65)

  • M. Senti et al.

    Relationship of age-related myocardial infarction risk and Gln/Arg 192 variants of the human paraoxonase gene: the Regicor study

    Atherosclerosis

    (2001)
  • D.K. Sanghera et al.

    The codon 55 polymorphoism of the paraoxonase 1 gene is not associated with risk of coronary heart disease in Asian Indians and Chinese

    Atherosclerosis

    (1998)
  • M.I. Mackness et al.

    Serum paraoxonase activity in familial hypercholesterolaemia and insulin-dependent diabetes mellitus

    Atherosclerosis

    (1991)
  • B. Mackness et al.

    Serum paraoxonase (PON1) 55 and 192 polymorphism and paraoxonase activity and concentration in non-insulin dependent diabetes mellitus

    Atherosclerosis

    (1998)
  • B.N. Patel et al.

    Serum esterase activities and hyperlipidaemia in the streptozotocin-diabetic rat

    Biochim. Biophys. Acta

    (1990)
  • Z. Ahmed et al.

    Apolipoproteina AI promotes the formation of phosphatidylcholine core aldehydes that are hydrolysed by paraoxonase (PON1) during high density lipoprotein oxidation with a peroxynitrite donor

    J. Biol. Chem.

    (2001)
  • Z. Ahmed et al.

    Multiple substrates for paraoxonase 1 during oxidation of phosphatidylcholine by peroxynitrite

    Biochem. Biophys. Res. Commun.

    (2002)
  • S. Deakin et al.

    Enzymatically active paraoxonase 1 is located at the external membrane of producing cells and released by a high affinity, saturable, desorption mechanism

    J. Biol. Chem.

    (2002)
  • R.J. Brushia et al.

    Baculovirus-mediated expression and purification of human serum paraoxonase 1A

    J. Lipid. Res.

    (2001)
  • M.N. Oda et al.

    Paraoxonase 1 over-expression in mice and its effect on high density lipoproteins

    Biochem. Biophys. Res. Commun.

    (2002)
  • B.J. Van Lenten et al.

    Oxidised phospholipids induce changes in hepatic paraoxonase and Apo J but not monocyte chemattractant protein-1 via interleukin-6

    J. Biol. Chem.

    (2001)
  • P. Kleemola et al.

    Dietary determinants of serum paraoxonase activity in healthy humans

    Atheroslcerosis

    (2002)
  • M. Aviram et al.

    Human serum paraoxonase (PON1) is inactivated by oxidised low density lipoprotein and preserved by antioxidants

    Free Rad. Biol. Med.

    (1999)
  • S. Arrol et al.

    Vitamin E supplementation increases the resistance of both LDL and HDL to oxidation and increases cholesteryl ester transfer activity

    Atherosclerosis

    (2000)
  • O. Kordonouri et al.

    Modulation by blood glucose levels of activity and concentration of paraoxonase in young patients with type 1 diabetes mellitus

    Metabolism

    (2001)
  • B.J. Kudchodkar et al.

    Dietary fat modulates serum paraoxonase 1 activity in rats

    J. Nutr.

    (2000)
  • M.I. Mackness et al.

    Paraoxonase activity is reduced by a pro-atherosclerotic diet in rabbits

    Biochem. Biophys. Res. Commun.

    (2000)
  • M. Aviram et al.

    Pomegranate juice consumption reduces oxidative stress, atherogenic modification of LDL, and platelet aggregation: studies in humans and in atherosclerotic apolipoprotein E-deficient mice

    Am. J. Clin. Nutr.

    (2000)
  • M.S. van der Gaag et al.

    Daily moderate alcohol consumption increases serum paraoxonase activity; a diet controlled, randomised intervention study in middle-aged men

    Atherosclerosis

    (1999)
  • B. Mackness et al.

    Human serum paraoxonase

    Gen. Pharmacol.

    (1998)
  • M. Aviram et al.

    Atorvastatin and gemfibrozil metabolites, but not the parent drugs are potent antioxidants against lipoprotein oxidation

    Atherosclerosis

    (1998)
  • P.N. Durrington et al.

    Effects of two different fibric acid derivatives on lipoproteins, cholesteryl ester transfer, fibrinogen, plasminogen activator inhibitor and paraoxonase activity in type IIb hyperlipoproteinaemia

    Atherosclerosis

    (1998)
  • Cited by (0)

    View full text