Postprandial modulation of serum paraoxonase activity and concentration in diabetic and non-diabetic subjects

Nutrition, Metabolism & Cardiovascular Diseases (2006) 16, 457e465

www.elsevier.com/locate/nmcd

Postprandial modulation of serum paraoxonase activity and concentration in diabetic and non-diabetic subjects Sandra Beer a, Xenia Moren b, Juan Ruiz a, Richard W. James b,* a

Division of Endocrinology, Diabetes and Metabolism, University Hospital, Lausanne, Switzerland Clinical Diabetes Unit, Division of Endocrinology, Diabetes and Nutrition, University Hospital, Geneva, Switzerland

b

Received 30 June 2005; received in revised form 5 September 2005; accepted 7 September 2005

KEYWORDS Oxidative stress; Atherosclerosis; Lipoprotein; Hypertriglyceridaemia

Abstract Objectives: To analyse the HDL associated anti-oxidant enzyme paraoxonase-1, during postprandial hyperlipaemia. Methods and results: Type 2 diabetic patients (n Z 72), glucose intolerant patients (n Z 10) and controls (n Z 38) consumed a high fat:high carbohydrate meal. Blood samples were collected up to 4 h and analysed for lipids and paraoxonase-1. In vitro studies examined HDL function with respect to the enzyme. There were significant postprandial increases in serum triglycerides. Paraoxonase-1 activity decreased significantly throughout the postprandial phase. Concentrations of the enzyme initially decreased significantly, but returned to fasting concentrations at 4 h. Specific activities were significantly lower at 4 h, compared to fasting. The decrease in specific activity was linked to the dynamic phase of postprandial lipoprotein metabolism. Apo AI limited loss of paraoxonase-1. HDL isolated after being subjected to postprandial conditions in vitro had reduced capacity to associate with and stabilise PON1. Conclusions: Postprandial hyperlipaemia was associated with changes to serum paraoxonase-1, consistent with a reduced anti-oxidant potential of HDL. No differences were observed between diabetic and non-diabetic patients, suggesting that the effect was linked to postprandial hyperlipaemia. Modifications to paraoxonase-1 could contribute to increased risk of vascular disease associated with postprandial lipaemia, particularly in diabetic patients, who are already deficient in serum paraoxonase-1. ª 2005 Elsevier B.V. All rights reserved.

* Corresponding author. Service of Endocrinology, Diabetes and Nutrition, University Hospital, 24, rue Micheli-du-Crest, 1211 Geneva 14, Switzerland. Tel.: C41 22 372 93 04; fax: C41 22 372 93 09. E-mail address: [email protected] (R.W. James). 0939-4753/$ - see front matter ª 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.numecd.2005.09.005

458

S. Beer et al.

Introduction

Methods

Postprandial dyslipidaemia is a risk factor for cardiovascular disease [1]. It is of particular relevance to diabetic patients, as underlying metabolic anomalies linked to diabetes can exaggerate postprandial dyslipidaemia [2]. A number of pathological consequences arises from postprandial dyslipidaemia in type 2 diabetes, including an increase in oxidative stress and endothelial dysfunction [3,4]. The serum enzyme paraoxonase-1 (PON1) is considered to be one of the principal determinants of the anti-oxidant capacity of high density lipoproteins (HDL) [5]. This activity has led to the hypothesis that PON1 offers protection against atherosclerosis. In this context, absence of the enzyme has been unambiguously linked to lesion development in animal models [6,7], whilst data from human studies are consistent with reduced serum PON1 being associated with increased risk of coronary disease [8e10]. Moreover, certain patient subgroups at high risk of cardiovascular disease have reduced serum PON1, notably diabetic patients [11e14]. The ability to protect LDL from oxidation may be of particular relevance to diabetic patients as oxidized LDL has been implicated in endothelial dysfunction [3,15,16]. There is growing interest in factors that can modulate serum activity of PON1 [17]. Our recent studies have shown the importance of HDL, which greatly facilitates PON1 secretion [18]. Both the lipid and peptide components of HDL play a role, as phospholipids promote secretion of the enzyme whilst apolipoprotein (apo) AI stabilises PON1 after its association with HDL. In addition to a strong genetic influence [19], environmental factors also affect the enzyme. Amongst these, studies have shown that diet has an impact [20e24]. These reflect principally the longer-term impact of modulating diet composition. One group has reported on the enzyme in the postprandial phase, although this involved a fat meal with high content of oxidized lipids [25,26]. Oxidation products are known to inhibit PON1 activity [27]. Most studies have relied on activity alone to follow the enzyme, which may mask more fundamental changes in PON1 metabolism, given the susceptibility of enzyme activity to inactivation. The present study was designed to examine serum PON1 in the postprandial phase after ingestion of a high fat meal. The aims were to examine the hypothesis that postprandial lipaemia would influence serum PON1 and to compare the responses in type 2 diabetic patients and nondiabetic subjects.

Study population Type 2 diabetic and impaired fasting glucose (IFG) patients were recruited from the outpatient clinic, University Hospital, Lausanne. The diagnoses of type 2 diabetes and IFG were based on the 1998 WHO criteria. Healthy control subjects were matched for age and gender. The latter had fasting glycaemia !6.1 mmol/l, a BMI !30 and blood pressure !140/ 90. They had no history of heart, lung, kidney, endocrine or liver disease and were not taking any medication. The IFG subjects had the same characteristics as the healthy controls except for fasting glycaemia O6.1 mmol/l !7 mmol/l. The protocol was approved by the hospital Ethics Committee (Lausanne) and carried out in accordance with the principles of the declaration of Helsinki, as revisited in 2000. Written informed consent was obtained from all subjects. Diabetic patients were treated with insulin alone (n Z 26), oral anti-diabetics alone (n Z 30; 8 with metformin alone, 1 with thiazolidinedione alone, 4 with sulfonylurea alone, 29 with more than one oral anti-diabetic) or with both insulin and oral anti-diabetic drugs (n Z 16). Other medication included angiotensin converting enzyme inhibitor (n Z 46), statin (n Z 35), fibrate (n Z 5), and aspirin (n Z 40). The diabetic patients were routinely screened according to local guidelines for macroangiopathy and microangiopathy. In the presence of one or more cardiovascular risk factors, coronary artery disease was screened with either myocardial scintigraphy or stress test echocardiography, and if positive, a coronarography was performed. Peripheral vascular disease and cerebrovascular disease were screened for (history of claudication, stroke or with arterial Doppler of the legs and the carotid arteries if clinically relevant). Diabetic nephropathy was screened with the albumin:creatinine ratio calculated in a morning urinary spot [28] (considered positive if O2 mg/mmol) [18]. Diabetic retinopathy was screened by routine eye examination. Seven patients had macroangiopathy alone, 13 patients had microangiopathy alone, 17 had both macroand microangiopathy and 35 were free of vascular complications.

Meal All participants reported fasting at 8 a.m. The diabetic patients were asked not to take their usual medication, including insulin, the last dose

PON-1 and postprandial hyperlipaemia being administered the previous day. Waist/hip ratio, BMI, blood pressure and heart rate were measured. An intravenous cannula was inserted into an antecubital vein to allow serial blood sampling and a first sample was collected in the fasting state. The study subjects then consumed a high fat, high carbohydrate meal (lipids (45% fat cream) 358 g; carbohydrates (corn flakes, saccharose) 397 g; proteins (45% fat cream) 33 g for a lipid:carbohydrate:protein ratio of 45:50:4. Blood was sampled in the fasting state and 1, 2, 3 and 4 h postprandially. All blood (serum) samples were centrifuged and stored at ÿ20  C.

459 [28]. PON1 serum activities and concentrations were quantified as described [29].

Statistical analyses ANOVA and repeated measures ANOVA, including between and within subject interaction analyses were used to evaluate the time effect (postprandial period) with post hoc analyses (Turkeye Kramer) where relevant. PON1 activities measured during in vitro studies were compared by paired and unpaired Student’s t-tests. Analyses were performed using the JMP statistical package (SAS Institute Inc., Cary, NC).

Incubation of serum with lipoprotein lipase Aliquots (2.0 ml) of serum from pools of healthy donors were incubated at 37  C, with and without purified lipoprotein lipase (Sigma-Aldrich, Fluka, Switzerland, 31 mg/ml, 2 h). Artificial triglyceride/ phospholipid vesicles (Lipovenous, Fresenius Kabi, Switzerland) were added to raise triglyceride concentrations to those observed during the meal (2.5e3.0 mM). After incubation, sodium chloride (1.5 M) was added to inhibit lipoprotein lipase activity, the enzyme was added to tubes previously incubated without lipase, and HDL isolated by ultracentrifugation at a density between 1.063 and 1.21 g/ml [28]. HDL was dialysed against phosphate buffered saline and stored under nitrogen. Prior to use, HDL was treated with 1 mM EDTA to inhibit irreversibly endogenous PON1 activity, and the EDTA subsequently removed by dialysis.

PON1 release from cells and enzyme stability HDL-stimulated PON1 release from cells was analysed as described previously [18]. Briefly, Chinese hamster ovary cells stably transfected with human PON1 were incubated in serum free medium containing HDL at 25 mg/ml (HDL phospholipid concentration). HDL concentration was defined in terms of phospholipid content as it is the primary determinant of the ability of the lipoprotein to stimulate PON1 release [18]. Conditioned medium was collected after 2 h and 7 h to analyse enzyme activity. Aliquots of conditioned medium were then stored at 4  C or 37  C and activity measured after 16 h to determine enzyme stability [18].

Biochemical analyses Serum lipids and lipoproteins were analysed by automated procedures, as described previously

Results Clinical and biochemical parameters of the subgroups are given in Table 1. The controls and diabetic patients were of equivalent age. Significant differences were observed across the subgroups for a number of variables. With respect to lipid variables, differences in cholesterol and LDLcholesterol were largely due to higher concentrations in glucose intolerant patients compared to the other two subgroups. Conversely differences for HDL-cholesterol and apo AI reflected lower concentrations in diabetic patients compared to the other subgroups. Fasting triglyceride concentrations rose from the control group through glucose intolerant patients to type 2 diabetic patients (Table 1). PON1 activities and concentration were lower in diabetic patients (Table 1). The differences were of the same order of magnitude as those reported in other studies [13,14], the lack of significance probably reflecting lower patient numbers. Glucose intolerant patients had similar PON1 values to non-diabetic participants (Table 1). Fig. 1A shows serum triglyceride concentrations after the test meal in each of the sub-groups. The highly significant differences observed for fasting triglycerides between controls and the diabetic/ glucose intolerant groups were maintained in the postprandial phase. In addition, increases in total serum concentrations of triglycerides in each subgroup were highly significant (p ! 0.0001 for all groups). Conversely, the postprandial incremental rises in serum triglycerides (Fig. 1B) were not significantly different between the sub-groups, with the exception of glucose intolerant patients. There was a tendency, however, for diabetic/glucose intolerant patients to show greater increases in postprandial triglycerides. No significant differences in PON1 variables were observed between diabetic patients when examined

460 Table 1

S. Beer et al. Clinical and biochemical characteristics of the population

Parameter

Type 2

IFG

Controls

ANOVA

n (M/F) Age (years) BMI HbA1C Fasting glycaemia Triglycerides Cholesterol LDL-cholesterol HDL-cholesterol Apo AI PON1 activity Arylesterase Paraoxonase Concentration

72 (55/17) 56.2 G 7.8 30.9 G 5.1*,# 7.5 (6.2e9.8) 7.6 (5.4e11.6)*,# 1.89 G 0.42* 4.89 G 0.84# 2.71 G 0.74# 1.28 G 0.33* 1.23 G 0.25*

10 (8/2) 55.2 G 9.1 25.2 G 2.9

38 (25/13) 56.0 G 11.1 23.0 G 2.5

0.44 0.92 !0.0001

5.8 (5.2e6.3) 1.62 G 0.49* 5.83 G 0.71* 3.64 G 0.66* 1.44 G 0.30 1.38 G 0.24

4.8 (4.2e5.4) 0.94 G 0.47 5.01 G 0.96 2.96 G 0.89 1.61 G 0.34 1.44 G 0.25

!0.0001 !0.0001 0.008 0.002 !0.0001 !0.001

89.7 G 23.6 146 (21e402) 74.8 G 23.1

97.1 G 26.6 273 (50e660) 79.2 G 24.5

96.2 G 24.3 233 (73e490) 79.0 G 24.6

0.34 0.34 0.67

Lipids are in mmol/l (mean G SD), apolipoproteins in g/l (mean G SD), arylesterase in U/ml (G SD), fasting glycaemia in mmol/l (median 90% CI), HbA1c in percent (median 90% CI), and paraoxonase in (activity) U/ml (median 90% CI), concentration (mg/ ml G SD). *Significantly different from controls; #Significantly different from glucose intolerant subgroup.

as a function of the presence of diabetic complications (data not shown). Forty-eight percent of the diabetic group were on statin treatment, but no significant interaction between statin and the postprandial measures was observed for PON1 variables or triglycerides. There were no correlations between enzyme activities or PON1 mass and measures of glycaemic control (HbA1c (measured only in diabetic patients) or fasting blood glucose). BMI was negatively correlated with activity (arylesterase; r Z ÿ0.26, p ! 0.02) and mass (r Z ÿ0.10, p Z 0.28), which may reflect lower PON1 in diabetic patients who had a significantly higher BMI. Enzyme activity was positively correlated with HDL-

cholesterol (arylesterase, r Z 0.27, p ! 0.01) and apo AI (arylesterase, r Z 0.25, p ! 0.005), as was PON1 mass. There were no significant associations between fasting triglyceride concentrations and either PON1 activities or serum concentrations. Initial analyses of the consequences of postprandial dyslipidaemia were examined in the sub-groups individually. No significant differences in the responses were noted between the sub-groups, as attested by no significant between group interactions by repeated measures ANOVA. The data were pooled for subsequent analyses. In the pooled data, a highly significant difference in activity (arylesterase) was observed

2

5

B

4

Triglycerides (mmol/L)

Triglycerides (mmol/L)

A

3

2

1

1

0

-1

0 t0

t2

Time (h)

t4

t2 - 0

t4 - 0

t4 - 2

Time (h)

Figure 1 Serum triglyceride concentrations before and after a high fat meal. (A) Serum triglyceride concentrations. (B) Incremental rises in serum triglycerides. Open squares, Type 2 diabetic patients; black diamonds, IFG patients; open circles, controls.

PON-1 and postprandial hyperlipaemia Table 2

461

PON1 activities and concentration in the fasting and postprandial phase

Parameter

t0

t2

t4

MANOVA

Arylesterase Paraoxonase Mass Specific activity (U/mg) Arylesterase Paraoxonase

92.7 G 24.1#,* 164 (173e431) 76.7 G 23.5

85.6 G 21.8 157 (71e406) 72.4 G 24.7

79.8 G 22.1 159 (72e396) 76.7 G 25.4

!0.001 0.08 !0.001

1.26 G 0.31* 2.91 G 1.56

1.26 G 0.32* 2.90 G 1.56

1.10 G 0.31 2.75 G 1.40

!0.001 0.006

Arylesterase is in U/ml (mean G SD), paraoxonase is in U/ml (median (90%CI)), mass is in mg/ml (mean G SD). #Significantly different from t2; *Significantly different from t4.

through the postprandial period (Table 2), reflecting a sustained decrease in activity from t0 to t4. Paraoxonase activity similarly showed a decrease in the postprandial period, but it did not attain statistical significance (p Z 0.08). Serum concentrations of PON1 also showed highly significant changes in the postprandial period (Table 2). Concentrations initially paralleled changes in activity, with a decrease at 2 h, but they subsequently diverged from activity measures, and rose to reach fasting concentrations at the 4 h postprandial time. In consequence, specific activities (activity/ unit mass PON1) were equivalent for arylesterase and paraoxonase at fasting and 2 h, but were significantly lower at the 4 h postprandial time point (Table 2). The latter was underlined by analysing the correlations between serum triglyceride concentrations and specific activities. In fasting serum, there was no significant correlation between these parameters (triglycerides t0 vs arylesterase specific activity t0, r Z 0.01, p Z 0.88). However, at the 4 h postprandial time point, serum triglycerides were significantly, negatively correlated with arylesterase specific activity (r Z ÿ0.23, p Z 0.017). There was also a negative correlation between the absolute postprandial rise

in triglycerides (that is, t4et0) and changes to arylesterase specific activity (r Z ÿ0.20, p Z 0.039). A second analysis focused on the divergent changes in paraoxonase activity and mass observed in the latter stages of hypertriglyceridaemia (between t2 h and t4 h) and its relationship to HDL. Here, the structural peptide apo AI was used as a marker of HDL. Although there were no significant group differences between apo AI concentrations at t2 h and t4 h (Fig. 2A), there was heterogeneity in the individual responses. As shown in Fig. 2B, changes in apo AI concentrations (t4et2) were positively correlated with changes to arylesterase activities (r Z 0.22, p ! 0.02). That is, a rise in apo AI from t2 to t4 was associated with a less important fall in PON1 activity. The rise in apo AI during t2 to t4 was also associated with an increase in serum concentrations of PON1 (r Z 0.21, p ! 0.04). A further series of in vitro studies examined the consequences of postprandial changes to HDL for PON1 metabolism (Fig. 3 shows typical results from 1 of three independent studies). These revealed that HDL-stimulated release of PON1 from cells was significantly lower (p ! 0.0001) for HDL isolated from hypertriglyceridaemic, lipase-treated

0.15

A

0.05

ARE (t4-t2)

Change (g/L)

0.1

0 -0.05 -0.1 -0.15 t2 - t0

t4 - t0

Postprandial period

t4 - t2

25 20 15 10 5 0 -5 -10 -15 -20 -25 -30 -35 -40 -45

B

-0.4

-0.2

0

0.2

0.4

Apo AI (t4 - t2)

Figure 2 (A) Changes in apo AI serum concentration in the postprandial period. (B) Correlation between changes in serum apo AI concentrations between 2 h and 4 h postprandial samples and changes in arylesterase activity.

462

S. Beer et al.

Discussion

15

Activity (OD/100ul)

12

9

6

HDL-HTG-37

HDL-Cont-37

HDL-HTG-7h

HDL-Cont-7h

HDL-HTG-2h

HDL-Cont-2h

3

Figure 3 Arylesterase activities in culture medium conditioned with CHO cells transfected with human PON1. Serum free medium containing HDL at 25 mg/ml (HDL phospholipid concentration) was incubated for 2 h or 7 h with cells and removed to analyse enzyme activity. Aliquots were then incubated 16 h at 4  C or 37  C to determine the stability of released PON1. HDL-Cont-2 h, HDL from non-treated serum after 2 h incubation with cells; HDL-HTG-2 h, HDL from hypertriglyceridaemic lipase-treated serum after 2 h incubation with cells; HDL-Cont-7 h, HDL from non-treated serum after 7 h incubation with cells; HDL-HTG-7 h, HDL from hypertriglyceridaemic lipase-treated serum after 7 h incubation with cells; HDL-Cont-37, control HDL conditioned medium after incubation at 37  C, 16 h; HDL-HTG-37, hypertriglyceridaemic, lipase-treated HDL conditioned medium after incubation at 37  C, 16 h.

serum for both short-term (2 h) and longer term (7 h) incubations with cells (a mean 13% reduction compared to control HDL (Fig. 3)). Stability of secreted PON1 was analysed by comparing arylesterase activity in conditioned medium incubated at 37  C to those maintained at 4  C (Fig. 3). There was no loss of activity from control HDL (mean 0.95 G 3.2% loss). Conversely, HDL from hypertriglyceridaemic, lipase-treated serum lost significantly more activity (9.6 G 1.8%, p ! 0.004 vs control). There were also changes in the composition of isolated HDL. Compared to control HDL, HDL isolated from hypertriglyceridaemic serum treated with lipoprotein lipase had lower phospholipid:protein (0.54 vs 0.61) and phospholipid: cholesterol (0.90 vs 1.10) concentration ratios. Lipoprotein lipase added to serum or HDL containing endogenous (non-inhibited) PON1 activity did not modulate such activity (data not shown).

Postprandial dyslipidaemia is considered to be an important contributor to the risk of atherosclerotic disease [1]. This may involve several mechanisms, including accumulation of atherogenic lipolytic remnants [30], deleterious changes to LDL and HDL structure [31] and a negative effect on endothelial function [32]. The present study suggests another mechanism whereby postprandial dyslipidaemia could influence atherogenic processes, by modulating serum paraoxonase. Given the antioxidant role attributed to PON1, a reduction in its activity/concentration would be particularly deleterious in the postprandial phase, notably in diabetic patients where there is evidence of increased oxidative stress [3]. A second novel and intriguing observation of the present study is the divergent impact of postprandial hypertriglyceridaemia on activity and peptide mass. Initially (up to 2 h) both decreased, suggesting that loss of activity was due to loss of PON1 protein. In this context, specific activities were comparable at t0 and t2 h. Subsequently, fasting concentrations of PON1 protein were recovered but in a less active form (significantly lower specific activity). There are a number of possible explanations for this phenomenon. The enzyme can be inactivated by oxidation products [27], and postprandial, alimentary lipids are able to provide such a source [33]. However, it is not apparent why this should lead to a simultaneous loss of PON1 protein. Alternatively, HDL composition/metabolism in the postprandial phase is in a dynamic state due to on-going catabolism of triglyceride-rich lipoproteins [34,35]. Such modifications may accelerate loss of associated PON1, which could explain the simultaneous loss of activity and protein. As for recovery of peptide, but not activity, we recently demonstrated a mechanism for hepatic secretion of PON1, which involved its desorption from the cell surface by a suitable acceptor complex, nominally HDL [18]. The study also demonstrated that the composition of HDL can modulate its capacity to stabilise enzyme activity [18]. To examine this possibility we isolated HDL from sera subjected in vitro to simulated postprandial conditions (lipoprotein lipase mediated lipolysis of hypertriglyceridaemic serum). The data indicate that the lipoprotein had (i) a reduced capacity to stimulate PON1 release from cells (as suggested by the fact that both short and long incubation periods were associated with reduced PON1 activity) and (ii) a lesser ability to stabilise enzyme activity. These in vitro studies are

PON-1 and postprandial hyperlipaemia consistent with our hypothesis that postprandial conditions can modulate HDL function with regard to PON1. In this context, the data in Fig. 2 concur with the importance of HDL metabolism on PON1, as postprandial changes to apo AI would appear to limit loss of activity and facilitate recovery of PON1 peptide. A final consideration arises from our recent studies showing that triglyceride-rich lipoproteins can also act as acceptors for PON1 release from hepatocytes [36]. The postprandial rise in triglyceride-rich lipoproteins may compete with HDL for PON1 released from hepatocytes. PON1 associated with triglyceride-rich lipoproteins is, however, less stable, which may contribute to the reduced specific activity that we observe. We hypothesise that it is not triglyceride concentrations per se, which modulate serum PON1, but the changes occurring in the postprandial phase. Thus, in fasting serum, triglycerides are not correlated with specific activity. However, higher triglyceride concentrations in the postprandial phase (4 h) are associated with a significant reduction in specific activity. It would suggest that the dynamic phase of postprandial dyslipidaemia impacts on PON1 metabolism, as illustrated by in vitro data using isolated HDL. Overall, the results underline the need to limit postprandial rises in triglycerides to minimize their impact on PON1, a recommendation that coincidentally is proposed to lower coronary risk. A limitation of the present study is that the antioxidant potential of postprandial HDL was not determined. It should be noted, however, that changes in PON1 concentration were observed, which must affect its anti-oxidant capacity. Although the changes were limited, in a previous investigation, also in diabetic patients, we showed that decreases in PON1, of the magnitude observed in the present study, were sufficient to diminish the anti-oxidant capacity of HDL [13]. There is some controversy concerning the extent to which enzyme activity with exogenous substrates reflects anti-oxidant capacity, although dose-dependent decreases in enzyme activity, by specific inhibition, were associated with a dosedependent reduction in anti-oxidant potential [37]. A minimal conclusion, however, is that the PON1 is sensitive to changes occurring during the postprandial phase. There have been few studies of the behaviour of paraoxonase in the postprandial phase. Oxidized lipids were shown to down-regulate hepatic PON1 mRNA [38], but it is unlikely to be a factor in the present study due to the rapidity with which serum PON1 changes are observed. Sutherland et al. [25] reported lower postprandial PON1 activity in healthy people fed used cooking oil. Oxidation

463 products in the used oil were proposed as the cause of the decrease in PON1 activity. No data were available on PON1 protein levels to indicate whether the changes reflected inhibition of the enzyme or loss of protein. We observed no significant differences in the postprandial PON1 response between diabetic and non-diabetic subjects. However, given that PON1 serum levels have been consistently shown to be lower in diabetic patients, the postprandial decrease in PON1 would accentuate this underlying PON1 deficiency. As such patients are also subject to a greater degree of oxidative stress [39], postprandial changes to PON1 would appear more deleterious in a diabetic setting. In conclusion, the present study is the first to suggest that serum PON1 metabolism is subject to modulation in the presence of postprandial hypertriglyceridaemia. It involves loss of PON1 protein, and although serum protein concentrations are recovered at 4 h, it is a less active form of PON1. Further studies are necessary to evaluate the contribution of postprandial modifications to HDL as a factor in changes to PON1. Alterations to PON1 activity have implications for the protective role of the enzyme against environmental toxins. The observed modifications would also reduce the anti-oxidant potential of HDL during a phase of lipoprotein metabolism that has been linked to increased risk of atherosclerosis. The changes appear to be particularly deleterious for diabetic patients, as they accentuate a deficiency in serum PON1.

Acknowledgements The study was supported by grants from the Swiss National Science Foundation (No 31-64788.01 and 31-105310.04 to RWJ and No 32-65308.01 to JR), the Swiss Cardiology Foundation (to RWJ) and the Swiss Diabetes Foundation (to JR).

References [1] Karpe F. Postprandial lipoprotein metabolism and atherosclerosis. J Intern Med 1999;246:341e55. [2] Chen YD, Swami S, Skowronski R, Coulston A, Reaven GM. Differences in postprandial lipemia between patients with normal glucose tolerance and noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1993;76:172e7. [3] Ceriello A, Taboga C, Tonutti L, Quagliaro L, Piconi L, Bais B, et al. Evidence for an independent and cumulative effect of postprandial hypertriglyceridemia and hyperglycemia on endothelial dysfunction and oxidative stress

464

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

S. Beer et al. generation: effects of short- and long-term simvastatin treatment. Circulation 2002;106:1211e8. Ceriello A, Quagliaro L, Piconi L, Assaloni R, Da Ros R, Maier A, et al. Effect of postprandial hypertriglyceridemia and hyperglycemia on circulating adhesion molecules and oxidative stress generation and the possible role of simvastatin treatment. Diabetes 2004;53:701e10. Watson AD, Berliner JA, Hama SY, La Du BN, Faull KF, Fogelman AM, et al. Protective effect of high density lipoprotein associated paraoxonase. Inhibition of the biological activity of minimally oxidised low density lipoprotein. J Clin Invest 1995;96:2882e91. Shih DM, Gu L, Xia Y-R, Navab M, Li W-F, Hama S, et al. Mice lacking serum paraoxonase are susceptible to organophosphate toxicity and atherosclerosis. Nature 1998;394: 284e7. Shih DM, Xia YR, Wang XP, Miller E, Castellani LW, Subbanagounder G, et al. Combined serum paraoxonase knockout/apolipoprotein E knockout mice exhibit increased lipoprotein oxidation and atherosclerosis. J Biol Chem 2000;275:17527e35. Jarvik GP, Rozek LS, Brophy VH, Hatsukami TS, Richter RJ, Schellenberg GD, et al. Paraoxonase (PON1) phenotype is a better predictor of vascular disease than is PON1(192) or PON1(55) genotype. Arterioscler Thromb Vasc Biol 2000;20:2441e7. Leviev I, Righetti A, James RW. Paraoxonase promoter polymorphism T(-107)C and relative paraoxonase deficiency as determinants of risk of coronary artery disease. J Mol Med 2001;79:457e63. Mackness B, Davies GK, Turkie W, Lee E, Roberts DH, Hill E, et al. Paraoxonase status in coronary heart disease: are activity and concentration more important than genotype? Arterioscler Thromb Vasc Biol 2001;21:1451e7. Abbott CA, Mackness MI, Kumar S, Boulton AJ, Durrington PN. Serum paraoxonase activity, concentration, and phenotype distribution in diabetes mellitus and its relationship to serum lipids and lipoproteins. Arterioscler Thromb Vasc Biol 1995;15:1812e8. Paragh G, Seres I, Balogh Z, Varga Z, Karpati I, Matyus J, et al. The serum paraoxonase activity in patients with chronic renal failure and hyperlipidemia. Nephron 1998; 80:166e70. Boemi M, Leviev I, Sirolla C, Pieri C, Marra M, James RW. Serum paraoxonase is reduced in type 1 diabetic patients compared to non-diabetic, first degree relatives; influence on the ability of HDL to protect LDL from oxidation. Atherosclerosis 2001;155:229e35. Boemi M, Sirolla C, Testa R, Cenerelli S, Fumelli P, James RW. Smoking is associated with reduced serum levels of the antioxidant enzyme, paraoxonase, in Type 2 diabetic patients. Diabet Med 2004;21:423e7. Pinkney JH, Stehouwer CD, Coppack SW, Yudkin JS. Endothelial dysfunction: cause of the insulin resistance syndrome. Diabetes 1997;46(Suppl. 2):S9e13. Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol 2003;23:168e75. Durrington PN, Mackness B, Mackness MI. The hunt for nutritional and pharmacological modulators of paraoxonase. Arterioscler Thromb Vasc Biol 2002;22:1248e50. Deakin S, Leviev I, Gomaraschi M, Calabresi L, Franceschini G, James RW. 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;277:4301e8.

[19] Leviev I, James RW. Promoter polymorphisms of the human paraoxonase PON1 gene and serum paraoxonase activities and concentrations. Arterioscler Thromb Vasc Biol 2000; 20:516e21. [20] Shih DM, Gu L, Hama S, Xia Y-R, Navab M, Fogelman AM, et al. Genetic-dietary regulation of serum paraoxonase expression and its role in atherogenesis in a mouse model. J Clin Invest 1996;97:1630e9. [21] Kudchodkar BJ, Lacko AG, Dory L, Fungwe TV. Dietary fat modulates serum paraoxonase 1 activity in rats. J Nutr 2000;130:2427e33. [22] Kleemola P, Freese R, Jauhiainen M, Pahlman R, Alfthan G, Mutanen M. Dietary determinants of serum paraoxonase activity in healthy humans. Atherosclerosis 2002;160:425e32. [23] de Roos MNG, Schouten EG, Scheek LM, van Tol A, Katan MB. Replacement of dietary saturated fat with trans fat reduces serum paraoxonase activity in healthy men and women. Metabolism 2002;51:1534e7. [24] Sierksma A, van der Gaag MS, van Tol A, James RW, Hendriks HF. Kinetics of HDL cholesterol and paraoxonase activity in moderate alcohol consumers. Alcohol Clin Exp Res 2002;26:1430e5. [25] Sutherland WH, Walker RJ, de Jong SA, van Rij AM, Phillips V, Walker HL. Reduced postprandial serum paraoxonase activity after a meal rich in used cooking fat. Arterioscler Thromb Vasc Biol 1999;19:1340e7. [26] Wallace AJ, Sutherland WH, Mann JI, Williams SM. The effect of meals rich in thermally stressed olive and safflower oils on postprandial serum paraoxonase activity in patients with diabetes. Eur J Clin Nutr 2001;55:951e8. [27] Aviram M, Rosenblat M, Billecke S, Erogul J, Sorenson R, Bisgaier CL, et al. Human serum paraoxonase (PON1) is inactivated by oxidised low density lipoprotein and preserved by antioxidants. Free Rad Biol Med 1999;26:892e904. [28] James RW, Pometta D. Differences in lipoprotein subfraction composition and distribution between type 1 diabetic men and control subjects. Diabetes 1990;39:1158e64. [29] Blatter Garin M-C, Abbot C, Messmer S, Mackness MI, Durrington P, Pometta D, et al. Quantification of human serum paraoxonase by enzyme-linked immunoassay: population differences in protein concentrations. Biochem J 1994;304:549e54. [30] Karpe F, Boquist S, Tang R, Bond GM, de Faire U, Hamsten A. Remnant lipoproteins are related to intimamedia thickness of the carotid artery independently of LDL cholesterol and plasma triglycerides. J Lipid Res 2001;42:17e21. [31] Deckelbaum RJ, Oschry Y, Rose L, Eisenberg S. Plasma triglyceride determines structure-composition in low and high-density lipoproteins. Arteriosclerosis 1984;4: 225e31. [32] Maggi FM, Raselli S, Grigore L, Redaelli L, Fantappie S, Catapano AL. Lipoprotein remnants and endothelial dysfunction in the postprandial phase. J Clin Endocrinol Metab 2004;89:2946e50. [33] Staprans I, Pan X-M, Rapp JH, Feingold KR. Oxidized cholesterol in the diet is a source of oxidized lipoproteins in human serum. J Lipid Res 2003;44:705e15. [34] Patsch JR, Karlin JB, Scott LW, Smith LC, Gotto Jr AM. Inverse relationship between blood levels of high density lipoprotein subfraction 2 and magnitude of postprandial lipemia. Proc Natl Acad Sci U S A 1983;80:1449e53. [35] James RW, Pometta D. Postprandial lipemia differentially influences high density lipoprotein subpopulations LpAI and LpAI. AII. J Lipid Res 1994;35:1583e91.

PON-1 and postprandial hyperlipaemia [36] Deakin S, Moren X, James RW. Very low density lipoproteins provide a vector for secretion of paraoxonase-1 from cells. Atherosclerosis 2005;179:17e25. [37] Aviram M, Rosenblat M, Bisgaier CL, Newton RS, PrimoParma SL, La Du BN. Paraoxonase inhibits high-density lipoprotein oxidation and preserves its functions. A possible peroxidative role for paraoxonase. J Clin Invest 1998;101: 1581e90.

465 [38] Van Lenten BJ, Wagner AC, Navab M, Fogelman AM. Oxidized phospholipids induce changes in hepatic paraoxonase and ApoJ but not monocyte chemoattractant protein-1 via interleukin-6. J Biol Chem 2001;276: 1923e9. [39] Ceriello A. New insights on oxidative stress and diabetic complications may lead to a ‘‘causal’’ antioxidant therapy. Diabetes Care 2003;26:1589e96.