In 59 healthy human subjects (37 male and 22 female) the concentrations of the lipid-soluble antioxidants alpha- and gamma-tocopherol, alpha- and beta-carotene, lycopene, cryptoxanthin, canthaxanthin, and lutein + zeaxanthin were determined in plasma (mumol/L) and in isolated low density lipoproteins (LDL) (mumol/mmol cholesterol). Plasma alpha-tocopherol concentrations were significantly correlated with plasma total cholesterol concentrations (r2 = 0.51, P < 0.0001) yet not with the LDL alpha-tocopherol content (r2 = 0.05, ns). Plasma gamma-tocopherol concentrations were weakly correlated with plasma total cholesterol (r2 = 0.12, P < 0.003) and both absolute and cholesterol standardized plasma gamma-tocopherol concentrations correlated strongly with the LDL gamma-tocopherol content (r2 = 0.58 and r2 = 0.72, respectively). In contrast, carotenoid concentrations did not correlate with cholesterol concentrations, but their LDL content correlated significantly with the respective plasma concentrations (r2 = 0.67 to 0.92, all P < 0.0001). In a subgroup of study subjects (n = 13) the distribution of vitamin E and carotenoids among LDL was calculated. The proportion of plasma alpha- and gamma-tocopherol found in LDL was 48 +/- 7 (range, 36-61%) and 41 +/- 7%, respectively, suggesting that LDL was in most of these subjects not the main carrier for these antioxidants. The lipophilic carotenoids, however, were predominantly carried by LDL (e.g., beta-carotene: 87 +/- 10%), whereas the proportion of the more polar ones carried by LDL was much smaller (e.g., lutein + zeaxanthin: 36 +/- 6%). The results of this study show that plasma alpha-tocopherol concentrations are not predictive for the alpha-tocopherol content of LDL in nonsupplemented individuals. This finding could have implications in interpreting the cause of the inverse relationship between plasma alpha-tocopherol and risk of atherosclerosis.