In quails and frogs, carotenoids are accumulated as yellowish oil droplets in the cone inner segments or RPE, respectively,
39 40 41 42 whereas in primates carotenoids are mostly concentrated in the Henle fiber layer.
16 Because of this pattern of accumulation, it can be argued that quail and frog may not serve as appropriate models for investigating the role of carotenoids in prevention of diseases such as macular degeneration in humans. Nonetheless, the data presented herein clearly indicate that these animals are likely to possess similar families of enzymes that may be responsible for the biochemical transformation of dietary (3
R,3′
R,6′
R)-lutein and (3
R,3′
R)-zeaxanthin in humans. Based on several human supplementation studies involving purified lutein, zeaxanthin, and lycopene, we have hypothesized a series of metabolic reactions that may explain the conversion of these dietary carotenoids to their presumed metabolites in humans.
43 44 In accordance with these findings and the data presented herein, the possible transformations for dietary lutein and zeaxanthin in humans, quails, and frogs may be summarized as a series of oxidation-reduction and double-bond isomerization reactions, as shown in
Figure 6 . The allylic oxidation of dietary lutein can result in the formation of (3
R,6′
R)-3-hydroxy-β,ε-carotene-3′-one (3′-oxolutein), which may undergo reduction either to revert back to dietary lutein or else epimerize at C-3′ to form 3′-epilutein. Both of these nondietary carotenoids have been identified in most ocular tissues of humans
(Table 1) . Alternatively, stereospecific double-bond isomerization of dietary zeaxanthin can also result in the formation of 3′-epilutein that may subsequently undergo allylic oxidation to yield 3′-oxolutein. Because 3′-oxolutein and 3′-epilutein are found in human plasma, it is not known whether their presence in ocular tissues is due to their transport through the circulatory system or whether these carotenoids may be formed locally in the eye by an independent process. However, the data presented in
Table 1 provides the most compelling evidence for metabolic conversion of dietary (3
R,3′
R,6′
R)-lutein to (3
R,3′
S;
meso)-zeaxanthin in the human ocular tissues through double-bond isomerization. This is because no detectable concentration of (3
R,3′
S;
meso)-zeaxanthin is found in human plasma and liver, despite the presence of this carotenoid in human eye tissues. Unfortunately, the dietary history and plasma carotenoid profile of the subject whose liver was analyzed in the current study is not known. Nonetheless, high levels of dietary lutein and zeaxanthin were present in this liver sample, whereas no (3
R,3′
S;
meso)-zeaxanthin could be detected. The highest concentration of (3
R,3′
S;
meso)-zeaxanthin relative to dietary (3
R,3′
R)-zeaxanthin was found in human macula and RPE-choroid
(Table 1) where carotenoid metabolism would be expected. The conversion of dietary lutein and zeaxanthin to their presumed metabolites as depicted in
Figure 6 , may be light induced and/or enzymatically mediated. Bone et al.
4 have demonstrated that the ratio of (3
R,3′
R)-zeaxanthin to (3
R,3′
S;
meso)-zeaxanthin for regions of human retina is subject to the choice of the area used in sectioning the retina. This ratio was shown by these investigators to reach approximately 1:1 if a 3-mm diameter punch of macula is selected and analyzed. We have chosen a 5-mm section of the macula and have found that the ratio of (3
R,3′
R)-zeaxanthin to (3
R,3′
S;
meso)-zeaxanthin is nearly 3:1. Therefore, selection of a larger diameter of the macula for the analysis of zeaxanthins may explain the difference between our data and those reported by Bone et al.
4