Specific antioxidant activity (SAA) (i.e., activity related to the molar or gallic acid equivalent amount of antioxidant) of natural polyphenolic mixtures or pure phenolic compounds was studied using their capacity to delay the conjugated diene production brought about by in vitro LDL copper-mediated or AAPH-mediated oxidation. The cinnamic acid series (caffeic, sinapic, ferulic acids) displayed a constant SAA over a large range of concentrations, whereas the benzoic acid series (gallic and protocatechuic acids) showed much higher SAA at low concentrations. The natural phenolic mixtures had a constant SAA. The highest SAA was obtained with caffeoyl esters (caffeoylquinic, rosmarinic, and caffeoyltartaric acids) and catechin for the copper-oxidation and the AAPH-oxidation system, respectively. Phenolic mixtures and acids delayed vitamin E depletion and decreased proinflammatory lysophosphatidylcholine production. As with polyphenols, probucol delayed lysophosphatidylcholine and conjugated dienes production, at higher concentrations, but was not effective at preventing vitamin E depletion. Polyphenols prevent the oxidation of LDL and its constituents (vitamin E, phosphatidylcholine), which is compatible with an antiinflammatory and antiatherosclerotic role in pathophysiological conditions.

The purpose of this double clinical study was (1) to evaluate the effect of one single intake (300 ml) of red wine (RW) on the plasma antioxidant capacity (pAOC) and plasma phenolics over the 24-h time period following the intake, and (2) to compare the long-term effects of daily intakes (250 ml/d) of RW, white wine (WW) and Champagne (CH) on the plasma and LDL characteristics of healthy subjects. In the first part, blood samples were collected just before and after wine consumption. In the second part, subjects received the 3 types of wine successively, only at the mealtime, over 3-week periods separated by a 3-week wash out. Blood samples were drawn in fasting condition before and after each 3-week wine consumption period. The peak of pAOC was at 3-4 h following the single intake of RW, that of catechin was at 4 h (0.13 micromol/l) and that of gallic acid and caffeic acid was earlier (< or = 1.5 and 0.3 micromol/l, respectively). In plasma, the major form of gallic acid was 4-O-methylated, but a minor form (the 3-O-methyl derivative) appeared. In the long term study, no wine was able to change LDL oxidizability, but some other parameters were modified specifically: RW decreased pAOC (without changing TBARS and uric acid plasma levels), LDL lipids and total cholesterol (TC), and increased plasma apoA1, whereas CH increased plasma vitamin A. The beneficial effect of RW seems to mainly be explained by its action on lipid and lipoprotein constants, and not by its antioxidant one.

Oxidized low density lipoprotein (LDL) plays an important role in atherogenesis. It is generally thought that LDL is mainly oxidized in the intima of vessel walls, surrounded by hydrophilic antioxidants and proteins such as albumin. The aim of this study was to investigate the possible interrelationships between oxidation resistance of LDL and its protein and lipid moieties. Proteins and to a lesser extent lipids, appeared to be the major determinants in the LDL Cu2+-oxidation resistance, which in turn depend on the ultracentrifugation (UC) procedure used. Comparing high speed/short time (HS/ST, 4 h), high speed/long time (HS/LT, 6-16h) and low speed/long time (LS/LT, 24h) conditions of UC, HS with the shortest time (4h) led to prepare LDL (named LDL.HS-4 h) with higher total protein and triglyceride contents, unchanged total cholesterol, phospholipids and Vitamin E, and higher Cu2+-oxidation resistance. Among proteins, only albumin allows to explain changes. PAF acetyl hydrolase appeared to be unaffected, whereas its pro-oxidant role was established and found only in the absence of albumin. In contrast the pro-oxidant role of caeruloplasmin took place regardless of the albumin content of LDL. The antioxidant effect of albumin (the oxidation lag time was doubled for 20mol/mol albumin per LDL) is assumed to be due to its capacity at decreasing LDL affinity for Cu2+. Interestingly, the LDL.HS-4 h albumin content mirrored the intrinsic characteristics of LDL in the plasma and was not affected by added free albumin. Moreover, it has been verified that in 121 healthy subjects albumin was the best resistance predictor of the Cu2+-oxidation of LDL.HS-4 h, with a multiple regression equation: lag time (min) = 62.1 + 0.67(HSA/apoB) + 0.02(TG/apoB)-0.01(TC/apoB); r = 0.54, P< 0.0001. Accounted for by lag time, the oxidation resistance did not correlate with alpha-tocopherol and ubiquinol contents of LDL. The mean albumin content was about 10mol/mol, and highly variable (0-58 mol/mol) with subjects. The LDL.HS-4h may account for the status of LDL in its natural environment more adequately than LDL resulting from other conditions of UC.