There is growing need for accurate information regarding the bioavailability of carotenoids, both with respect to carotenoidsper se and to the vitamin A value of provitamin A carotenoids in foods or supplement preparations. Little quantitative information is currently available, owing primarily to the lack of adequate methods to assess carotenoid bioavailability. Methods applied to xenobiotic drugs are in most cases not useful for carotenoids, many of which circulate in appreciable quantities in human plasma. Reported ranges of carotenoid bioavailability (% dose absorbed) range from 1-99, and variability is generally high both within and between treatments. With the current methods, relative bioavailability is more readily assessed than absolute bioavailability. The most commonly applied methods include measuring the increase in plasma carotenoid concentration following chronic intervention, and use of postprandial chylomicron (PPC) carotenoid or retinyl ester response following a single dose of carotenoid. The advantages and limitations of these approaches, together with examples of each, are discussed. A new PPC approach utilizing extrinsic-stable-isotope-labelled vitamin A (2H4-labelled retinyl acetate) is under development in our laboratory, and examples of its application are presented. The currently available data suggest that oil solutions of carotenoids are more bioavailable than those from food matrices, and heating can improve the bioavailability of carotenoids from some food products. Increased availability of labelled carotenoids and retinoids should aid the development of reliable methods of carotenoid bioavailability assessment. Such data are needed for dietary recommendations, supplement formulation, and design of intervention strategies involving carotenoids. Bioavailability : Carotenoids: Vitamin A: Postprandial chylomicron techniqueCarotenoids are lipid-soluble plant pigments common in photosynthetic plants, animal tissues, and in the bloodstream and tissues of human subjects consuming plant and animal products. These pigments are of physiological interest in human nutrition, since some are vitamin A precursors and many exhibit radical or singlet oxygen trapping activity (Miller et al. 1996), and as such have potential antioxidant effects in vivo. Consequently, it is of value to determine the bioavailability of carotenoids from both foods and supplement forms in human subjects, and identify and quantify those factors which significantly affect carotenoid uptake and tissue disposition. Reliable data are needed to formulate effective policies for combating vitamin A deficiency and prevention of chronic degenerative diseases. At the present time there is considerable debate over such issues as the actual vitamin A value offoods containing provitamin A carotenoids, and the relative importance of the large number of factors which may potentially influence carotenoid bioavailability. It is currently not possible to assign with confidence specific values to the absorption efficiency of a specific caro...

The extent to which processing affects the carotene or vitamin A value of foods is poorly understood. An extrinsic reference method was used to estimate the mass of carotenes and vitamin A derived from various preparations made from the same lot of carrots. Using a repeated-measures design, nine healthy adult subjects consumed test meals of either carrot puree (commercial baby food) or boiled-mashed carrots on separate days; six of the subjects also consumed a test meal of raw-grated carrot. Test meals supplied 34.7 micromol (18.6 mg) carrot beta-carotene (beta C), plus 6 micromol deuterium-labeled retinyl acetate (d(4)-RA) in oil solution. Baseline-adjusted carotene and retinyl ester (R-ester) area-under-curve (AUC) responses in the triacylglycerol-rich lipoprotein (TRL) fraction (0-8.5 h) were determined using HPLC and gas chromatography-mass spectrometry. The masses of absorbed beta C, alpha-carotene (alpha C) and R-ester were estimated by comparing their AUC values with that of deuterium-labeled retinyl ester (d(4)-R-ester), assuming the latter represented 80% of the d(4)-RA reference dose. Absorption of beta C and alpha C was approximately twofold greater from carrot puree than from boiled-mashed carrots, whereas the retinol yield was only marginally (P = 0.11) influenced by treatment. Carotene and R-ester absorption from raw-grated carrot was intermediate to, and did not differ significantly from the cooked preparations. The vitamin A yield (puree, 0.53 mg; boiled-mashed, 0.44 mg) of cooked carrot containing 18.6 mg beta C was substantially less than that predicted by current convention and limited primarily by intestinal carotene uptake. Processing can therefore significantly improve bioavailability of carrot carotenes, and in some cases influence the carotene value more than the intrinsic vitamin A value.

Absorption and metabolism of [13C]9-cis-beta-carotene ([13C]9c beta C) was studied in three subjects after a single oral dose. Subjects given 1.0 mg [13C]beta-carotene (mean: 99.4% 9-cis-beta-carotene, 0.6% all-trans-beta-carotene; dose A) had substantial concentrations of [13C]all-trans-beta-carotene ([13C]tr beta C) and [13C]all-trans retinol ([13C]retinol) but very low concentrations of [13C]cis-beta-carotene ([13C]cis beta C) in saponified plasma 5 h after dosing, as determined by HPLC and isotope-ratio mass spectrometry. There was no evidence of appreciable absorption of [13C]9-cis retinol. To determine the proportion of [13C]tr beta C and [13C]retinol derived from [13C]9c beta C, a second set of studies in the same subjects was performed with the same isomeric composition except with 13C labeling only in all-trans-beta-carotene (dose B). The results indicated that > 95% of plasma [13C]tr beta C and [13C]retinol observed after dose A was derived from [13C]9c beta C. The concentrations of [13C]tr beta C observed, in excess of that derived from the trace amounts of [13C]tr beta C in the dose, indicated that a significant proportion of the [13C]9c beta C dose was isomerized to [13C]tr beta C before entering the bloodstream. Although precise quantitative estimates of the extent of isomerization of 9-cis-beta-carotene could not be made, it is apparent that cis-trans isomerization of 9-cis-beta-carotene to all-trans-beta-carotene contributed to the near absence of postprandial plasma 9-cis-beta-carotene after its oral administration in humans. The observation of different ratios of beta-carotene to retinol between the two dosing protocols suggests that isomerization did not occur exclusively before uptake by the intestinal mucosa. These results indicate that isomerization of ingested 9-cis-beta-carotene before its secretion into the bloodstream limits the potential supply of 9-cis retinoids to tissues, and increases the vitamin A value of 9-cis-beta-carotene.

Human lung type II cell derived A549 epithelial cancer cells and HepG2 hepatocytes constitutively express cytochrome P4504F2, a P450 we previously identified as a tocopherol-omega-hydroxylase. To determine if A549 cells would metabolize tocochromanols via the omega-hydroxylase pathway, we compared the metabolism of tocopherols (alpha-, gamma-, delta-TOH) and tocotrienols (alpha-, gamma-, delta-T3) in these 2 cell lines. Cultures were incubated with alpha-, gamma-, or delta-TOH, or the analogous T3s, and synthesis of their metabolites quantitated by GC-MS. A549 cells metabolized all tocochromanols 2-3 times more extensively than HepG2 cells (P < 0.001) except alpha-TOH, a difference not related to cell uptake of substrate but rather was reflective of greater microsomal TOH-omega-hydroxylase enzyme activity. Notably, 9'-carboxychromanols were the major metabolites of all gamma- and delta-TOHs and T3s in A549 cultures, whereas 3'- and 5'-carboxychromanols predominated in HepG2 cultures. Accumulation of 9'-carboxychromanols in A549 cultures was due to their inefficient conversion to 7'-carboxychromanols relative to HepG2 cells. Sesamin inhibited tocochromanol metabolism in both cells types, and neither cell type exhibited evidence of alternative (sesamin-insensitive) pathways of metabolism. TOH-omega-hydroxylase activity was undetectable in rat primary lung type II cells, suggesting that expression of activity was associated with transformation of normal type II cells to cancer cells. Long-chain carboxychromanol metabolites of gamma-TOH and other forms of vitamin E can be biosynthesized in A549 cultures for assessment of their biological activity, including their potential inhibition of synthesis of inflammatory mediators.

The triacylglycerol-rich lipoprotein and d4-RA method, which controls for variation in chylomicron kinetics in vivo and RE recovery during analysis, is useful for obtaining quantitative estimates of the vitamin A potential of single meals.