Original contribution
Dietary hydroxy fatty acids are absorbed in humans: implications for the measurement of ‘oxidative stress’ in vivo

https://doi.org/10.1016/S0891-5849(01)00780-8Get rights and content

Abstract

Lipid peroxidation products formed in vivo or originating from the diet may lead to atherosclerosis. However, little is known about the absorption of these products in man. We studied the absorption of fat (30 g) containing 14–15 mg [U-13C]-labeled hydroxy or dihydroxy triglycerides in two groups of six apparently healthy women aged 40 ± 2 years. Post-prandial 13C-labeled hydroxy fatty acid concentration increased in a pattern somewhat different from that of plasma triglycerides, with peak levels being reached between 4 and 6 h. However, the amount of 13C-labeled oxidized fat absorbed (area under the curve of plasma concentrations from 0 to 8 h) was related to that of plasma triglycerides: 13C hydroxy vs TG (r = 0.88, p < .02), and 13C dihydroxy vs TG (r = 0.85, p < .05). 13C monohydroxy triglycerides appeared to be absorbed to a greater extent than those of 13C dihydroxy triglycerides. Although low levels of 13C hydroxy lipids could be detected in fasting plasma after 24 h, concentrations were very low. Dietary lipid oxidation products are absorbed. The measurement of hydroxy fatty acids in plasma total lipids may not be a valid marker of lipid peroxidation in vivo when subjects are not fasting.

Introduction

Oxidative modification of LDL may be critical in the development of atherosclerosis. This process is thought to occur solely within the arterial intima [1], where oxidized LDL induces monocyte chemotactic protein1 and scavenger receptors [2], [3], [4]. Monocytes and neutrophils can oxidize LDL [1], [2], [5]. Polyunsaturated fatty acids are highly susceptible to peroxidation and hydroperoxides, formed during cooking, can be converted to smaller organic compounds [6]. Hydroxy fatty acids, stable products derived from hydroperoxy fatty acids, are present in most seed oils and edible fats at concentrations of 1–10 μM [7]. They probably represent the major long-chain auto-oxidation products in edible fats.

It was generally assumed that oxidation products from oils and fats are poorly absorbed [6], [8]. Indeed animals can be resistant to the effects of abused oils in their diet [6], [8]. However, orally administered 14C linoleate oxidation products were incorporated into chylomicrons in rats. Oxidized lipids derived from chylomicrons were incorporated into very low-density lipoprotein particles [9]. The absorbed oxidation products were subsequently identified as hydroxy fatty acids [9].

Despite the importance of oxidized LDL in the development of atherosclerosis, little is known about absorption of dietary lipid oxidation products in humans. The plasma concentration of hydroxy plus hyderoperoxy fatty acids might be a valid marker of lipid peroxidation in vivo, provided these oxidation products are not absorbed. Suggestions that they are absorbed are based on nonspecific methods for analysis, such as thiobarbituric acid reactive substances or conjugated dienes [10], [11]. In the present study the absorption of stable isotope labeled hydroxy or dihydroxy fatty acids incorporated in triglycerides was studied using gas chromatography-mass spectroscopy in humans [12]. The advantage of using 13C-labeled hydroxy fatty acids is that the contribution of lipid peroxidation products formed in vivo following a fat load is excluded.

Section snippets

Volunteers

Twelve apparently healthy female volunteers (aged 40 ± 2 years) participating in a lipid-screening program of staff of the Royal Infirmary Edinburgh were recruited. All consumed their normal diet. Two were current smokers. The average body mass index (BMI) was 24.1 ± 1.1 kg/m2. One woman was overweight (BMI 27.1 kg/m2). Hypercholesterolemia (> 7 mmol/l or 270 mg/dl) and age (> 60 years) were exclusion criteria for this study. Fasting triglyceride and total cholesterol levels were within normal

Results

There was a 1.6-fold rise in plasma triglyceride concentrations following the consumption of 30 g fat (Fig. 1 and Table 1). Peak levels occurred between 4 and 6 h. The increase in naturally occurring hydroxy fatty acids in plasma reflected that of plasma triglycerides closely in 11/12 subjects (correlation coefficients ranging from 0.43–0.99, mean 0.81 ± 0.05).

The rise in the concentrations of [U-13C]-labeled hydroxy fatty acids showed a pattern somewhat different from that of plasma

Discussion

The plasma concentrations of [U-13C]-labeled hydroxy and dihydroxy fatty acids, determined by GC-MS, increased following their consumption in healthy women. Thus hydroxy and dihydroxy fatty acids are absorbed. There was a marked inter-individual variation in plasma [U-13C] hydroxy fatty acid concentrations and this reflected a similarly large variation in postprandial lipaemia in these women. The increase in plasma concentrations of [U-13C] hydroxy fatty acids was much higher than that of [U-13

Acknowledgements

R. Wilson was supported by a MAFF LINK Agro Food Programme grant AFQ112 (MAFF, Nestlé UK Ltd, Roche Products Ltd. and Van den Bergh Foods Ltd). R.A. Riemersma is supported by the British Heart Foundation. C.E. Fernie was supported by EC grant FAIR PL0594. We gratefully acknowledge the technical assistance from Margaret Millar and Claire Pearson. We wish to thank MAFF and the members of the Link management group (Nestlé UK Ltd, Roche Products Ltd. and Van den Bergh Foods Ltd) for their support.

References (27)

  • D Steinberg et al.

    Beyond cholesterol. Modifications of low density lipoprotein that increase its atherogenecity

    N. Engl. J. Med.

    (1989)
  • H.A Lehr et al.

    In-vitro effects of oxidized low-density-lipoprotein on cd11b cd18 and l-selectin presentation on neutrophils and monocytes with relevance for the in-vivo situation

    Am. J. Pathol.

    (1995)
  • M Klouche et al.

    Atherogenic properties of enzymatically degraded LDLselective induction of MCP-1 and cytotoxic effects on human macrophages

    Arterioscler. Thromb. Vasc. Biol.

    (1998)
  • Cited by (0)

    View full text