**3. Intestinal Absorption of Xanthophylls**

In addition to the bioaccessibility, carotenoid uptake by intestinal epithelial cells is also a critical factor for the carotenoid bioavailability. Only one part of the accessible carotenoid is taken up by the intestinal epithelial cells and secreted into lymph as chylomicrons for circulating in blood stream. After the chylomicrons are degraded by lipoprotein lipase, carotenoids in chylomicron remnants are taken up by the liver. The carotenoids are stored in liver or resecreted as very-low-density lipoprotein into the blood stream, and then delivered as low-density lipoprotein (LDL). Finally, carotenoids are taken up to tissues through the LDL receptor. Highly hydrophobic carotenoids such as Ά-carotene and lycopene are localized in the inner part of LDL, while less hydrophobic xanthophylls such as lutein and zeaxanthin are equally distributed to LDL and HDL, and localized in the outer surface area of the lipoprotein particles [28]. 

The intestinal absorption of carotenoids had been thought to be mediated by simple diffusion [29,30]. To characterize the human intestinal absorption of carotenoids, we compared the uptakes of various carotenoids by human intestinal Caco-2 cells [31]. The carotenoids solubilized at the same concentration in mixed micelles were incubated with the Caco-2 cells. The uptakes were correlated with their lipophilicity, suggesting that simple diffusion mediated the intestinal uptake of the carotenoids. The amounts of fucoxanthin and neoxanthin taken up by the cells were approximately 25% of that of lutein and were the lowest among the eleven carotenoids tested. These results indicated that fucoxanthin and neoxanthin were certainly absorbed in the Caco-2 cells, although the amounts were lower than that of lutein. 

In addition to the experiments using Caco-2 cells, we were able to confirm the absorption of these two xanthophylls in mice [32–34]. The xanthophylls solubilized in mixed micelles were orally administrated to male ICR mice. Fucoxanthinol and amarouciaxanthin A derived from fucoxanthin were detected in plasma and the liver [32,33]. A similar result was also reported in rats fed with fucoxanthin [35]. Neoxanthin and neochromes (formed from neoxanthin by intragastric acidity) were detected in plasma and the liver [34]. The plasma concentrations in the mice 2 h after administration of four purified carotenoids (40 nmol) in the independent experiments under almost the same condition were as follows: 36 nM for Ά-carotene [36]; 10 nM for lutein [36]; 35 nM for neoxanthin (neoxanthin and neochromes) [34] and 50 nM for fucoxanthin (fucoxanthinol and amarouciaxanthin A) [33]. Neoxanthin and fucoxanthin were confirmed to be absorbed at a similar level to those of Ά-carotene and lutein, and no selectivity for carotenoids tested was found in mice. 

In addition to rodents, it has been reported that fucoxanthin is absorbed in other animals such as tunicates [37,38], chicken [39], and aquatic insects [40]. However, fucoxanthin was not absorbed in freshwater fish [40]. East Asian people ingest fucoxanthin from foodstuffs such as sea squirt, sea urchins, mussel, and brown algae. However, no information on the absorption of fucoxanthin in humans has been available. Although neoxanthin and violaxanthin are ingested from green leafy vegetables under usual dietary habits, they were not found in human serum and milk by a detailed analysis of carotenoids [19]. Thus, it has been uncertain whether fucoxanthin and neoxanthin are absorbed in humans. 

We reported for the first time the bioavailability of fucoxanthin from edible brown algae (wakame) and of neoxanthin and violaxanthin from spinach in humans [41]. After the daily intake of stir-fried wakame containing 6.1 mg fucoxanthin for 1 week, the concentrations of fucoxanthin and its metabolites in plasma were analyzed by HPLC. Fucoxanthin and amarouciaxanthin A were not detected. Fucoxanthinol was detected, but the plasma concentration was under the quantification limit (1.0 nM). Similar to the case of fucoxanthin, the plasma concentrations of neoxanthin and violaxanthin after the intake of stir-fried spinach were under the quantification limit. On the other hand, both Ά-carotene and lutein, which were present with these epoxy xanthophylls in the same food matrix of spinach, were increased in the plasma [41], suggesting that little neoxanthin and violaxanthin in spinach were absorbed in humans. In contrast to the case of mice, selective absorption of carotenoids may occur in humans. 

The low bioavailability of these epoxy xanthophylls may be caused by their low bioaccessibility from spinach and wakame. However, the bioaccessibility of neoxanthin (neoxanthin and neochromes) from spinach was comparable with that of lutein and was greater than that of Ά-carotene in *in vitro* digestion study [34]. Similarly, the bioaccessibility of fucoxanthin from wakame was sufficiently 

high [41]. These results suggested that the bioaccessibility was not a limiting factor of the bioavailability. 

The absence of these epoxy xanthophylls in human plasma may be due to the rapid metabolism. However, the concentrations of these epoxy xanthophylls and their metabolites in the plasma were under the quantification limit even shortly after the intake of spinach and wakame [41], indicating that the rapid disappearance might not occur. 

It is possible that the level of these epoxy xanthophylls in plasma were estimated to be low due to unknown metabolic transformation such as hydrolysis of epoxide or formation of conjugates by detoxification enzymes after the intestinal uptake. For instance, fucoxanthinol 3ȝ-sulphate found in the egg yolk of hens fed with seaweed meal [39] might be formed from fucoxanthin in humans. 

The dietary water-soluble fibers, alginates in wakame may be associated with the low bioavailability of fucoxanthin from wakame in humans, because dietary watersoluble fibers inhibited the Ά-carotene and lutein uptake by Caco-2 cells [42]. Thus, it is necessary to reveal the bioavailability of isolated carotenoid to avoid the influence of the food matrix. 

There are several reports on the bioavailability of epoxy xanthophylls in the purified preparations and the oleoresins in human subjects. Oleoresin, which is extracted from plant materials, does not contain dietary fibers and any other polar substances. Capsanthin 5,6-epoxide and violaxanthin were not detected in chylomicron after ingestion of paprika oleoresin containing these epoxy xanthophylls [43]. However, 9-*cis* zeaxanthin, which was present at a lower amount than epoxy xanthophylls in paprika oleoresin, was found in chylomicron [43]. This result suggested that little capsanthin 5,6-epoxide and violaxanthin in paprika were absorbed in humans. Moreover, after a single oral dose of purified violaxanthin or lutein 5,6-epoxide suspended in corn oil, the two epoxy xanthophylls were not detected in the plasma [44]. In contrast, after an oral dose of purified Ά-carotene 5,6- epoxide (9.1 μmol) suspended in corn oil, the plasma concentration reached 2.29 μM [45]. Considering these experimental results with the oleoresin and purified xanthophylls, little epoxy xanthophylls that have higher polarity than Ά-carotene 5,6- epoxide would be absorbed by humans, consistent with the results of our human study  using spinach and wakame. The chemical structures of these epoxy xanthophylls are shown in Figure 1. 

To summarize the intestinal absorption of carotenoids, little of highly polar epoxy xanthophylls such as neoxanthin and violaxanthin were absorbed in humans independent of the food matrix. Fucoxanthin was absorbed in mice and several other animals, but not in humans and freshwater fishes. A selective absorption mechanism for carotenoids would be present in humans, but not in mice. Moreover, the selectivity in the intestinal absorption of carotenoids appears to differ among animal species. 
