The Carotenoid Pigment Zeaxanthin—A Review



ABSTRACT:  Scientific evidence linking several diseases with diet has brought to light the beneficial effects of a number of natural food ingredients. Zeaxanthin is one such natural pigment emphasized for its critical role in the prevention of age-related macular degeneration (AMD), the leading cause of blindness. The review highlights zeaxanthin as a carotenoid pigment with promising nutraceutical implications, and enumerates the important plant and microbial sources for its production, the absorptive pathway of zeaxanthin in human system, and methods to assess its bioavailability besides other relevant aspects.


Carotenoids are pigments naturally occurring in a number of fruits and vegetables. They are synthesized by all photosynthetic organisms and many nonphotosynthetic bacteria and fungi. They are liposoluble tetraterpenes originating from the condensation of isoprenyl units, which form a series of conjugated double bonds constituting a chromophoric system (Britton 1995). There are 2 main classes of naturally occurring carotenoids: (1) carotenes such as β-carotene and α-carotene, which are hydrocarbons, are either linear or cyclized at one or both ends of the molecule, and (2) xanthophylls, the oxygenated derivatives of carotenes. All xanthophylls produced by higher plants, such as violaxanthin, antheraxanthin, zeaxanthin, neoxanthin, and lutein, are also synthesized by green algae (Eonseon and others 2003). Epidemiological studies have established an inverse relationship between the risk of laryngeal, lung, and colon cancers and the consumption of foods containing carotenoids (Block and others 1992; Steinmetz and Potter 1993).

The chemical name of zeaxanthin is (all-E)-1,1′-(3,7,12,16-tetramethyl-1,3,5,7,9,11,13,15,17-octadecanonaene-1,18-diyl) bis [2,6,6-trimethylcyclohexene-3-ol]. Synonyms are: 3R, 3′R-β,β-carotene-3,3′-diol; all-trans-β-carotene-3,3′-diol; (3R,3′R)-dihydroxy-β-carotene; zeaxanthol; and anchovyxanthin. Zeaxanthin, the principal pigment of yellow corn, Zeaxanthin mays L. (from which its name is derived), has a molecular formula of C40H56O2 and a molecular weight of 568.88 daltons. Its CAS number is 144-68-3. It is composed of 40 carbon atoms, yellow in color, and naturally found in corn, egg yolks, and some of the orange and yellow vegetables and fruits such as alfalfa and marigold flowers (Nelis and DeLeenheer 1991; Handelman and others 1999; Humphries and Khachik 2003). Zeaxanthin exhibits no vitamin A activity. Zeaxanthin and its close relative lutein (Figure 1 and 2) play a critical role in the prevention of age-related macular degeneration (AMD), the leading cause of blindness (Snodderly 1995; Moeller and others 2000). Zeaxanthin is isomeric with lutein; the 2 carotene alcohols differ from each other just by the shift of a single double bond so that in zeaxanthin all double bonds are conjugated. Zeaxanthin is used as a feed additive and colorant in the food industry for birds, swine, and fish (Hadden and others 1999). The pigment imparts a yellow coloration to the skin and egg yolk of birds, whereas in pigs and fish it is used for skin pigmentation (Nelis and DeLeenheer 1991).

Figure 1—.

Structure of zeaxanthin.

Figure 2—.

Structure of lutein.

Stereoisomers of Zeaxanthin

Zeaxanthin has 2 chiral centers and, hence, 22 or 4 stereoisomeric forms. One chiral center is the number ‘3’ atom in the left end ring, while the other chiral center is the number ‘3’ carbon in the right end ring (Garnett and others 1998). One stereoisomer is (3R, 3′R)-zeaxanthin; the other is (3S-3′S)-zeaxanthin. The 3rd stereoisomer is (3R, 3′S)-zeaxanthin and the 4th (3S-3′R)-zeaxanthin. However, since zeaxanthin is a symmetric molecule, the (3R, 3′S)—and (3S, 3′R)—stereoisomers are identical. Therefore, zeaxanthin has only 3 stereoisomeric forms. The (3R, 3′S)—or (3S, 3′R)—stereoisomer is called meso-zeaxanthin. The principal natural form of zeaxanthin is (3R, 3′R)-zeaxanthin. (3R, 3′R)-zeaxanthin and meso-zeaxanthin are found in the macula of the retina, with much smaller amounts of (3S, 3′S)-zeaxanthin. Meso-zeaxanthin is a rare isomer present in significant quantities in commercially produced chickens and eggs in Mexico where it is commonly added to the feed to achieve desirable coloration in these products (Bone and others 2007).

Properties of Zeaxanthin

One gram of zeaxanthin dissolves in about 1.5 L of boiling methanol. The pigment is almost insoluble in petroleum ether and hexane. Its solubility in ether, chloroform, carbon disulphide, and pyridine is somewhat greater. Zeaxanthin dissolves in concentrated sulfuric acid with a fairly stable deep blue coloration. On treating a solution of the pigment in chloroform with antimony trichloride, a blue coloration is produced (Euler and others 1930).

Zeaxanthin is a polyene-like molecule, which contains 9 alternating conjugated carbon double and single bonds. The carbon backbone is terminated at each end by an ionone ring to which a hydroxyl group is attached. When excited with monochromatic laser light, it exhibits characteristic wavelength shifts of inelastically back-scattered light caused by vibrational modes in its chemical structure. Two characteristic carotenoid peaks shown in Figure 3 originate from rocking motions of the carbon–carbon single bond stretch vibrations (1159 cm−1) and from the carbon–carbon double bond stretch vibrations (1525 cm−1) of the molecule backbone (Bhosale and others 2003).

Figure 3—.

Resonance Raman spectra of zeaxanthin (ZX) extracted from bacteria and dissolved in methanol and of Flavobacterium multivorum culture broth (CB, plotted on right Y axis (Bhosale and others 2003).

The conjugated double-bond system constitutes the light-absorbing chromophore that gives carotenoids their attractive color and provides the visible absorption spectrum that serves as a basis for their identification and quantification. Cis-isomerization of a chromophore's double bond causes a slight loss in color, small hypsochromic shift, and hypochromic effect, accompanied by the appearance of a cis peak in or near the ultraviolet region. All-trans isomers absorb strongly in the visible region between 400 and 500 nm while cis-isomers exhibit absorption in the near-UV region, around 320 nm (Rodriguez-Amaya 2001). The visible spectrum of zeaxanthin, a derivative of β-carotene, resembles that of β-carotene. The ultraviolet and visible absorption data of zeaxanthin are shown in Table 1 with calculation of % III/II as indication of spectral fine structure (% III/II × 100) illustrated in Figure 4. Figure 5 shows the absorption spectra of isomers of zeaxanthin.

Table 1—.  Ultraviolet and visible absorption data of zeaxanthin (Davies 1976; Britton 1995).
CarotenoidSolventλ max nma% III/IIb
  1. aParentheses indicate a shoulder.

  2. bRatio of the height of the longest-wavelength absorption peak, designated III, and that of the middle absorption peak, designated II, taking the minimum between the 2 peaks as baseline multiplied by 100 (

ZeaxanthinAcetone(430) 452 479 
Chloroform(433) 462 493 
Ethanol(428) 450 47826
Petroleum ether(424) 449 47625
Figure 4—.

Calculation of % III/II as indication of spectral fine structure (% III/II × 100) (

Figure 5—.

Absorption spectra of the isomers of zeaxanthin. Spectra were recorded from pigment solution in MTBE-methanol (5:95, by volume) (Justyna and Gruszecki 2005).

Carotenoid molecules are strong Raman scatterers. Hence, nondestructive resonance Raman spectroscopy could be an extremely valuable method for the rapid quantitative assessment of carotenoids. There are reports of the detection of resonance Raman scattering of laser radiation of the carotenoid pigments from intact plant samples and fruit juices (Gill and others 1970).

Biosynthetic Pathway and Genetic Manipulation of the Pathway for Zeaxanthin Production

A paramount function of xanthophylls in all photosynthetic organisms, including cyanobacteria, is to provide protection against photooxidation. It is proposed that zeaxanthin protects the membrane directly against lipid peroxidation by reactive radicals that have been created as toxic byproducts during photosynthetic reactions. Another mechanism suggests specific xanthophylls to be involved in the de-excitation of singlet chlorophyll (1Chl) that accumulates in the light-harvesting complexes (LHC) under conditions of excessive illumination (Demmig-Adams 1990; Demmig-Adams and Adams 1992; Demmig-Adams and others 1996).

Carotenoid biosynthesis leads to the all-trans-forms. Hence, all-trans-lycopene, -lutein and -zeaxanthin predominantly occur in fresh fruits and vegetables. All-trans-isomers have therefore been assumed to be the thermodynamically stable form of carotenoids (Miebach and Behsnilian 2006). Carotenoids are synthesized from the basic C5-terpenoid precursor, isopentenyl diphosphate (IPP) and dimethylallyl pyrophosphate (DMAPP). IPP is converted into geranylgeranyl phosphate (GGPP) and the dimerization of GGPP leads to phytoene (7,8,11,12,7′,8′,11′,12′-octahydro-γ,γ-carotene) and the stepwise dehydrogenation via phytofluene (15 Z, 7, 8,11,12, 7′,8′-hexahydro-γ, γ- carotene), zeta carotene (7,8,7′,8′-tetrahydro-γ-γ-carotene), and neurosporene (7,8-dihydro-γ, γ-carotene) gives lycopene. Subsequent cyclization, dehydrogenation, and oxidation lead to hundreds of different natural carotenoids (Figure 6). In Flavobacterium R1529, nicotine blocks zeaxanthin biosynthesis by specifically inhibiting the cyclization reaction (McDermott and others 1974). Lycopene and rubixanthin replace zeaxanthin as the main carotenoid. In the absence of nicotine, lycopene is converted to β-carotene under anaerobic conditions and into zeaxanthin in the presence of oxygen.

Figure 6—.

Precursor compounds and major carotenes and xanthophylls in the carotenoid biosynthetic pathway in plants. DMAPP = dimethylallyl pyrophosphate; GGPP = geranylgeranyl pyrophosphate; IPP = isopentenyl pyrophosphate (Kopsell and Kopsell 2006).

The biosynthesis of IPP and DMAPP from acetyl-CoA via melavonate has been studied using animal cells and yeasts (Bochar and others 1999). Three acetate units afford the 5 carbon atoms of IPP from loss of 1 acetate carboxylic group as CO2. DMAPP is obtained from IPP by an isomerase. A mevalonate-independent 2nd pathway for the biosynthesis of IPP and DMAPP via 1-deoxy-D-xylulose 5-phosphate has been discovered in some eubacteria and plants (Rohmer and others 1993). The nonmevalonate pathway starts with the formation of 1-deoxyxylulose 5-phosphate from pyruvate and glyceraldehyde 3-phosphate catalyzed by 1-deoxyxylulose 5-phsophate synthase. The carbohydrate is converted into 2 C-methylerythritol 2, 4-cyclodiphosphate by a series of 4-reaction steps (Eisenreich and others 2002).

Genetic studies with Arabidopsis thalina (Rock and Zeevaart 1991; Rock and others 1994), Nicotiana plubaginifolia (Marin and others 1996), and the green alga Chlamydomonas reinhardtii (Niyogi and others 1997) revealed the presence of a single-gene coding for the zeaxanthin epoxidase enzyme. Thus, a single-gene product is apparently responsible for both the biosynthesis of violaxanthin during growth, and development and the epoxidation reaction leading to the return of zeaxanthin via antheraxanthin to violaxanthin following recovery after irradiance stress. Mutants with lesions in the zeaxanthin epoxidase gene not only are consequently deficient in antheraxanthin and violaxanthin, but also fail to synthesize neoxanthin (Niyogi and others 1997; Jin and others 2003). In addition, these mutants accumulate large amounts of zeaxanthin that are almost equivalent to the levels of violaxanthin found in the wild type, even when grown under nonstressed conditions. Mutations affecting zeaxanthin production exist in green algae, Scenedemus obliquus (Bishop and others 1995), Chlamydomonas reinhardtii (Niyogi and others 1997), and Dunaliella salina (Jin and others 2003).

In microalgae, the zeaxanthin content is regulated by light irradiance. When photosynthetic irradiance is greater than that required for the saturation of photosynthesis in the chloroplasts of plants and green algae, a reversible violaxanthin de-epoxidation reaction occurs to form antheraxanthin and, subsequently zeaxanthin, resulting in the accumulation of zeaxanthin in the chloroplast thylakoids. The enzyme that catalyzes this reaction, violaxanthin de-epoxidase, is localized in the lumen of the chloroplast thylakoids (Hager and Holocher 1994). When the absorbed irradiance is lower than that required for saturation of photosynthesis, zeaxanthin is converted back into violaxanthin by the enzyme zeaxanthin epoxidase with the monoepoxide antheraxanthin being an intermediate in this reversible oxidation–reduction process (Hager 1980). The xanthophyll cycle is thought to be essential for the protection of the photosynthetic apparatus from photooxidation, and zeaxanthin, antheraxanthin, and violaxanthin are associated with the light-harvesting complexes (Demmig-Adams and Adams 1992). In accordance with the role of zeaxanthin and violaxanthin in photosynthesis, ABA2mRNA (abscisic acid 2mRNA) is more abundant in photosynthetic than in nonphotosynthetic tissues.

ABA, a breakdown product of xanthophyll carotenoids (C40) via the C15 intermediate xanthoxin (Walton and Li 1995), modulates the growth and development of plants, particularly during seed formation and also in response to environmental stress (Zeevaart and Creelman 1988; Giraudat and others 1994). Mutants blocked in the early steps of carotenoid synthesis, for example, some viviparous mutants of maize (vp2, vp5, vp7, or vp9), lack carotenoids essential for photosystem protection and, therefore, exhibit photobleaching and ABA deficiency (Neill and others 1986). In contrast, mutants impaired in the downstream steps of carotenoid biosynthesis do not show photobleaching. The aba1 mutant of Arabidopsis and the aba2 mutant of Nicotiana plumbaginifolia are impaired in the epoxidation of zeaxanthin and have been shown to be either slightly or not at all affected in PSII photochemical efficiency (Rock and Zeevaart 1991; Rock and others 1992; Marin and others 1996; Tardy and Havaux 1996; Hurry and others 1997). Zeaxanthin is able to replace the missing epoxy-carotenoids, antheraxanthin, violaxanthin, and neo-xanthin as a stabilizing component of the light-harvesting complex II in the aba1 mutant of Arabidopsis.

The recent genetic elucidation of bacterial and plant carotenoid biosynthetic pathways leading to the accumulation of zeaxanthin, canthaxanthin, and astaxanthin may offer interesting alternatives for their in vivo production (Misawa and others 1995a, 1995b; Misawa and Shimada 1998). For instance, blue-green algae can be readily transformed with autonomously replicating plasmids, while endogenous genes can be disrupted by homologous recombination. A number of commercial possibilities have been proposed for recombinant blue-green algae (Lagarde and others 2000). Recently, Synecocystis sp. strain PCC 6830 was used as a transformation host to overproduce zeaxanthin in vivo. Moreover, the system developed in the study allowed for gene replacement without the introduction of antibiotic resistance cassettes in the final overexpressing strains. The absence of cassettes containing genes that confer antibiotic resistance in such strains is a positive feature highlighting the increasing desire of the biotechnology industry to avoid spreading antibiotic-resistant cassettes, thereby respecting the concerns of consumers and environmentalists.

A genetically engineered zeaxanthin-rich potato

In an attempt to provide a good supply of zeaxanthin in staple crops such as potatoes (Solanum tuberosum L.), 2 different potato varieties were genetically modified (Römer and others 2002). By transformation with sense and antisense constructs encoding zeaxanthin epoxidase, the conversion of zeaxanthin to violaxanthin was inhibited. Both antisense and cosuppression approaches yielded potato tubers with high levels of zeaxanthin. Depending on the transgenic lines and tuber development, zeaxanthin content was elevated by 4- to 130-fold, reaching values up to 40 μg/g dry weight.

Localization and Sources of Zeaxanthin

Xanthophylls are relatively hydrophobic molecules. Therefore, they are typically associated with membranes and/or noncovalently bound to specific proteins. In general, primary carotenoids are localized in the thylakoid membrane, while secondary carotenoids are found in lipid vesicles either in the plastid stroma or the cytosol. Most xanthophylls that are found in cyanobacteria and oxygenic photosynthetic bacteria are associated with chlorophyll (Chl)-binding polypeptides of the photosynthetic apparatus (Grossman and others 1995). Among nonphotosynthetic bacteria and, to a lesser extent, among photosynthetic bacteria and cyanobacteria, xanthophylls and their glycosides can be found in cytoplasmic and cell wall membranes where they are thought to influence membrane fluidity (Armstrong 1997).

Plant sources

Green leafy vegetables are good dietary sources of lutein, but poor sources of zeaxanthin. Dietary sources of zeaxanthin include yellow corn, orange pepper, orange juice, honeydew, mango, and chicken egg yolk. Jungalwala and Cama (1962) found zeaxanthin to comprise about 90% of the total carotenoids in the anthers of Delonix regia (Gul Mohr) flowers. Zeaxanthin is also the major carotenoid in cold-pressed marionberry, boysenberry, red raspberry, and blueberry seed oils, followed by β-carotene, lutein, and cryptoxanthin (Parry and others 2005). Zeaxanthin has also been identified in extracts from apricots, peaches, cantaloupe, and a variety of pink grapefruit (Ruby seedless) among carotenoids separated and quantified on C18 reversed-phase HPLC columns with low and high carbon loading (Khachik and others 1989). The major carotenoids of Viburnum tinus leaves are β-carotene and lutein (38.5% and 47%, respectively) with neoxanthin and zeaxanthin accounting for 7% and 6.8%, respectively (Chiralt and others 1990). Carotenoids identified in persimmon fruits are cis-mutatoxanthin, antheroxanthin, zeaxanthin, neolutein, cryptoxanthins, α-carotene, and β-carotene, and fatty acid esters of cryptoxanthin and zeaxanthin (Daood and others 1992). Zeaxanthin has also been identified in the skin, flesh, and oil of avocado (Ashton and others 2006). Lucerne contains 2% zeaxanthin and 40% lutein of the total xanthophyll content, while Fremontia (Sterculalareaceae) produces as much zeaxanthin as lutein (Goodwin 1976). The stereochemical correlation between capsanthin and zeaxanthin and the cooccurrence of the 2 pigments in Capsicum sp. have suggested a close biogenetic relationship between the two. Formation of capsanthin from zeaxanthin via antheraxanthin is indicated (Britton 1976). Of the total pigment content, zeaxanthin contributes to about 6.5%, 7.3%, and 15.9%, respectively, in the red, orange, and yellow varieties of Capsicum annuum (Goodwin 1976).

In corn, xanthophylls are mostly found in the horny endosperm. The total xanthophyll content is estimated to be 11 to 30 mg/kg (Zuber and Darrah 1987). In 1 study, yellow dent corn was found to contain a total xanthophyll content of 21.97 μg/g with lutein content of 15.7 μg/g, zeaxanthin content of 5.7 μg/g, and β-cryptoxanthin of 0.57 μg/g (Moros and others 2002). Commercial corn gluten meal has 7 times higher concentration of xanthophylls (145 μg/g), and deoiled corn contains 18 μg/g, indicating that the xanthophylls are probably bound to the zein fraction of corn proteins. Wolfberry (Lycium chinese), a small fruit used to improve vision in traditional Chinese medicine, contains concentrations of zeaxanthin dipalmitate that can approach 1 g/kg wet weight (Zhou and others 1999). The zeaxanthin concentration of fruits and vegetables is shown in Table 2 (Khachik and others 1989).

Table 2—.  Zeaxanthin concentration in fruits and vegetables.
Food itemsZeaxanthin, μg per 100 g
Beans (drained, green, canned) 44
Broccoli (drained, cooked, boiled without salt) 23
Carrots (raw) 23
Celery (drained, cooked, boiled, without salt)  8
Celery (raw)  3
Collards (drained, cooked, boiled, without salt)266
Corn (drained, sweet, yellow, canned, whole kernel)528
Lettuce (raw)187
Lettuce (raw) 70
Orange juice (frozen concentrate, unsweetened, diluted) 80
Oranges (all commercial varieties) 74
Papaya 20
Peaches (raw)  6
Peas (drained, green, canned) 58
Spinach (drained, cooked, boiled without salt)179
Spinach (raw)331
Squash, butternut280
Tangerines, raw (mandarin oranges)112

Microbial sources

Recently, there has been much interest in the microbial production of zeaxanthin (Ruther and others 1997; Jin and others 2003); however, unlike β-carotene and astaxanthin, very few microbes synthesize zeaxanthin as their predominant carotenoid (Johnson and Schroeder 1996).

Flavobacterium sp Among the sources of microbial xanthophylls, Flavobacterium sp. is reported to produce zeaxanthin as essentially its only carotenoid. The pigment formed by Flavobacterium consists of 95% to 99% zeaxanthin. Flavobacterium-produced zeaxanthin is identical to zeaxanthin from Zea mays (Gierhart 1995). Beta-carotene along with β-cryptoxanthin is known to act as precursors in the biochemical pathway of zeaxanthin production, and thus appreciable levels of these carotenoids (about 5% to 10%) were observed during initial growth phases of Flavobacterium sp. (Bhosale and others 2004). Hydroxylation of β-carotene and β-cryptoxanthin ultimately leads to accumulation of zeaxanthin.

However, for optimal industrial production, it is essential to continue strain improvement and to select better producers in order to maximize the yield. Currently, the only effective way to screen for better producers is to extract carotenoids from microbes and to perform HPLC analysis with ultraviolet/visible or photodiode array detectors. Improved yields of zeaxanthin may be obtained by culturing a microorganism of the genus Flavobacterium under conditions whereby the amounts of carbon and nitrogen present in the culture medium are maintained at a substantially constant ratio (Gierhart 1995).

Cultures of Flavobacterium sp. in a nutrient medium containing glucose or sucrose, sulfur-containing amino acids, such as methionine, cystine, or cysteine, pyridoxine, and bivalent metal ions selected from the group consisting of Fe2+, Co2+, Mo2+, or Mn2+ were able to produce up to 190 mg zeaxanthin/L, with a specific cell concentration of 16 mg/g dried cellular mass (Shepherd and others 1976). The optimized process claims to provide up to 500 mg of zeaxanthin/L culture at lower costs and more rapidly than known methods and microorganisms.

Several reports have shown different medium constituents other than environmental factors to affect zeaxanthin production. Among these factors, nitrogen and carbon sources play an important role in zeaxanthin production (Gierhart 1994; Alcantara and Sanchez 1999). For Flavobacterium, corn steep liquor is beneficial for pigment synthesis because of its richness in amino acids, minerals, and other factors necessary for growth (Goodwin 1971; Gierhart 1994). The effect on growth and production of zeaxanthin by Flavobacterium sp. was studied using different carbon and nitrogen sources in a chemically defined medium by Alcantara and Sanchez (1999). The best growth was supported by sucrose, and both asparagine and glutamine were found to stimulate growth and pigment formation. Carotenoid production and glucose consumption increased as a function of asparagine concentration. Flavobacterium sp. was found to utilize asparagine primarily as a nitrogen source for growth and production of zeaxanthin. In the presence of asparagine, high glucose concentrations decreased pigment production without affecting biomass formation. In the absence of glucose, asparagine could not support growth and zeaxanthin production (Table 3). Lactic acid and palmitic acid methyl esters are reportedly pigment promoters of zeaxanthin. With no special measures taken, fermentation of a Flavobacterium sp. in a medium containing glucose and corn steep liquor generates approximately 10 to 40 mg zeaxanthin/L (Nelis and DeLeenheer 1989). The yield increased to 335 mg/L by supplementation with palmitic esters, methionine, pyridoxine, ferrous salts by continuous addition of the nutrients, and reduction of temperature.

Table 3—.  Effect of different carbon and nitrogen sources on growth and zeaxanthin production (Alcantara and Sanchez 1999).
ConditionsGrowth (OD at 540 nm)Zeaxanthin (μg/mL)
  1. aFermentation was carried out for 48 h, 29 °C, and 180 rpm in CD medium supplemented with 20 mM NH4Cl.

  2. bFermentation was carried in CD medium supplemented with 100 mM glucose.

Carbon sourcesa
 Yeast extract (1%)2.38 ± 0.0700.75 ± 0.025 
 D-glucose (55 mM)1.55 ± 0.0660.15 ± 0.020 
 Sucrose (55 mM)1.82 ± 0.1020.15 ± 0.027 
 Xylose (55 mM)0.51 ± 0.0320.07 ± 0.017 
 No carbohydrate0.46 ± 0.0550.07 ± 0.011 
Nitrogen sources (7.5 mM)b
 L-asparagine3.74 ± 0.0500.38 ± 0.016 
 L-glutamine2.96 ± 0.1190.36 ± 0.011 
 L-cysteine1.89 ± 0.0900.07 ± 0.0009
 L-methionine1.85 ± 0.0150.08 ± 0.0095
 Glycine 0.6 ± 0.0200.03 ± 0.007 
 L-asparagine without glucose0.44 ± 0.1010.08 ± 0.009 
 L-asparagine without MgSO40.55 ± 0.0300.03 ± 0.006 
 L-asparagine + L-cysteine2.31 ± 0.0760.28 ± 0.003 

Another important factor in the production of microbial pigments is the oxygen provided to the culture medium (Goodwin 1971; Britton 1985). In general, an increase in oxygen supply to a highly active culture can increase its productivity. The growth of bacteria and fungi, measured in terms of dry weight, varies directly with the efficiency of aeration above a predetermined level depending on the available substrate. In addition to being required for growth in Flavobacterium, oxygen is also required for desaturation, cyclization, and oxygenation of carotenoids (Smith and Johnson 1958; Edwards 1985; Han and Mudgett 1992; Britton 1995). To improve zeaxanthin production by Flavobacterium, a fermentor with a high oxygen transfer rate is preferred. Agitation speed and aeration rate are the 2 factors that strongly affect the oxygen supply in stirred-tank fermentors. A proper combination of these factors can regulate the required oxygen supply to a growing culture. However, any additional supply of oxygen should be matched with the right availability of other nutrients such as corn steep liquor. Some factors that have an unexpectedly strong positive effect on β-carotene production at the expense of zeaxanthin formation should also be considered. In a study on carotenoid production by Flavobacterium multivorum ATCC 55238, urea and sodium carbonate were found to influence β-carotene production, which represented 70% (w/w) of the total carotenoid content (Bhosale and Bernstein 2004).

The nutrient media that are currently preferred for commercial-scale fermentation have eliminated corn flour and several other ingredients, and contain either high-maltose corn syrup of sugar beet molasses at concentrations ranging from 1% to 10% (w/v), along with corn steep liquor at 0.5% to 4% (w/v); ammonium sulfate heptahydrate at 0.5% (w/v); sodium chloride at 0.5% (w/v); magnesium sulfate heptahydrate at 0.1% (w/v); sodium acetate at 0.1% (w/v); ferrous sulfate heptahydrate at 0.001% (w/v); yeast extract at 0.2% (w/v); thiamine-HCl at 0.01% (w/v); between 1% and 6% (w/v) hydrolyzed casein; and vegetable oil at 1% (v/v) (Garnett and others 1998). Once the ingredients are mixed together, sufficient NaOH is added to raise the pH to 6.5. The culture medium is sterilized by autoclaving at 121 °C for 30 min, cooled to 27 °C, and inoculated with 5% to 10% (v/v) of a “liquid preculture” containing a strain of F. multivorum which produces the R-R isomer of zeaxanthin without producing S-S or S-R isomers of zeaxanthin, or any other carotenoids, in significant quantities.

Several strains of Flavobacterium have been taxonomically reevaluated as Paracoccus zeaxanthinifaciens sp. (Berry and others 2003). Metabolic engineering can be used to improve zeaxanthin production in the bacterium Paracoccus sp. strain PTA-3335, a mutant derived from a zeaxanthin-producing bacterium originally classified as a species of Flavobacterium (Schocher and Wiss 1975).

Erwinia herbicolaErwinia herbicola, a nonphotosynthetic bacterium, is yellow-colored due to accumulation of polar carotenoids, primarily mono- and diglucosides of zeaxanthin (Hundle and others 1992).

Neospongiococcum Among FDA-approved GRAS strains is Neospongiococcum, the only alga presently designated as GRAS for feeding poultry to enhance yellow pigmentation (21 C.F.R. section 73,275). The green alga Neospongiococcum excentricum is shown to produce up to 0.65% xanthophylls (dry mass basis) (Liao and others 1995).

Dunaliella salina A zeaxanthin-overproducing mutant strain zea1 generated from Dunaliella salina may be considered for commercial exploitation (Jin and others 2001, 2003). This mutant strain has a defect in the zeaxanthin-epoxidation step. Thus, the zea1 mutant lacks neoxanthin, violaxanthin, and antheraxanthin, but constitutively accumulates zeaxanthin in the thylakoid membrane even under normal growth conditions. Under normal growth conditions (low light), the mutant strain has 15-fold higher zeaxanthin content than the wild type. Previous efforts to generate zeaxanthin-overproducing E. coli strains using metabolic engineering have resulted in the production of 1.6 mg zeaxanthin/g dry weight; however, this value is just one-third of that produced by the zea1 strain of the photosynthetic microalga Dunaliella salina (6 mg zeaxanthin/g dry weight) (Albrecht and others 1999; Jin and others 2003).

Synechocystis sp Zeaxanthin is a natural constituet of the outer membrane of Synechocystis sp. PCC6714 (Jurgens and Weckesser 1985). In an attempt to increase zeaxanthin accumulation in a photoautotrophic prokaryote, Synechocystis sp. strain PCC 6803, a system designed to overexpress genes involved in carotenoid synthesis in the organism employs the psbAII gene, which encodes the highly expressed D1 protein of photosystem II and has a strong promoter in it. Synechocystis genome contains 3 genes coding for the D1 protein, psbAI, psbAII, and psbAIII, the latter two of which are expressed and can individually support normal photoautotrophic growth in the absence of the other two psbA genes (Mohamed and Jansson 1989). Thus, the psbAII locus can be used as an integration platform to overexpress genes in Synechocystis. Synechocystis sp. resulted in a 2.5-fold increase in zeaxanthin accumulation in the mutant strain (Lagarde and others 2000).

Microcystis aeruginosa The microalgae Microcystis aeruginosa is reported to produce the bioactive carotenoid zeaxanthin (Chen and others 2005).

Spirulina The blue-green alga Spirulina has zeaxanthin as one of its carotenoids. A marked enhancement in carapace color, an important quality parameter of cultured prawns, was observed when prawns were fed with Spirulina-supplemented diets. Zeaxanthin, one of the major carotenoids in Spirulina, was rapidly converted to astaxanthin (Liao and others 1993). Spirulina, also increases the yellowness and redness of broiled chicken due to accumulation of zeaxanthin within the flesh (Toyomizu and others 2001).

Phaffia rhodozyma Among yeasts, asporogenous Phaffia rhodozyma, although best known as an astaxanthin producer, is also reported to be a zeaxanthin producer (Hoshino and others 2004).

Zeaxanthinibacter enoshimensis is a zeaxanthin-producing marine bacterium of the family Flavobacteriaceae, isolated from the seawater off Enoshima island in Japan (Asker and others 2007a). Mesoflavibacter zeaxanthinifaciens is another novel zeaxanthin-producing marine bacterium of the family Flavobacteriaceae (Asker and others 2007b). The carotenoids of the red algae Corallina officinalis, C. elongate, and Jania sp. are reportedly composed of β-carotene, zeaxanthin, fucoxanthin, 9′-cis-fucoxanthin, fucoxanthinol, 9′-cis-fucoxanthinol, and 2 epimeric mutatoxanthins (Palermo and others 1991). The symbiotic blue-green algae Cyanophora paradoxa and Glaucocystis nostochinearum synthesize only β-carotene and zeaxanthin (Goodwin 1976). The thylakoid of the cyanobacteria Anacystis nidulans is reported to have zeaxanthin as one of its major carotenoids (Murata and others 1981). Dunaliella parva (Andrew and Britton 1990), Erythrotrichia carnea (Shlomai and others 1992), Dunaliella bardawil (Ben-Amotz and others 1982), Prochloron sp. (Withers and others 1978), and Pleurochloris commutata (Goodwin 1976) are some of the other microbial sources of zeaxanthin.

Extraction and Isolation of Zeaxanthin

The advantage of tetrahydrofuran (THF) in comparison to other organic solvents of Class 2 as an extraction solvent is that the carotenoid is highly soluble in the solvent, which allows for its efficient extraction from plant matrices (Khachik 2005). Solvents in Class 2 should be limited in pharmaceutical products because of their inherent toxicity. The choice of extracting solvents is based on a guideline set by the U.S. Dept. of Health and Human Services, Food and Drug Administration (FDA) in Docket No. 97D-0148 published in the Federal Register on May 2, 1997 (Vol 62, Nr 85, p 24301–9). The draft guideline recommends acceptable amounts of residual solvents in pharmaceuticals for the safety of patients as well as the use of less toxic solvents in the manufacture of drug substances and dosage forms.

Suitable modes of packaging and delivery for human ingestion include various forms such as lyophilized coarse-grain powder, viscous oily liquid, and micelles. An alternate form of packaging for human ingestion can utilize tablets, provided that a solid binder material is used which will hold the tablet together after manufacture. Such tablets may be coated with an enteric coating that remains intact until after the tablet passes through the stomach and enters the intestine. Zeaxanthin is believed to be susceptible to degradation at high pressures (Garnett and others 1998). Thus, it is assumed that compressed tablets will be less preferred or tablet binders that ensure good cohesion at lower pressures could be used.

FlavobacteriumspFlavobacterium multivorum, a nonfastidious and nonpathogenic bacterium that rapidly accumulates zeaxanthin, is considered to be a potential source of the carotenoid (Dasek and others 1973; Pasamontes and others 1997; Alcantara and Sanchez 1999; Masetto and others 2001). Synthesis of a pure form of the R-R isomer of zeaxanthin involves fermenting bacterial cells from a strain of Flavobacterium multivorum (ATCC 55238) (Garnett and others 1998). The bacterial strain can synthesize zeaxanthin as a sole carotenoid with virtually no substantial amounts of other carotenoids under proper fermentation conditions. Therefore, the extremely difficult task of purifying zeaxanthin by separating it from closely related carotenoids can be completely avoided if this bacterial strain or its descendants are used.

In the isolation and pure cultivation of the yellow Flavobacterium microorganisms, material from the natural source is suspended in physiological saline (Gierhart 1994). Streak cultures are applied to Petri dishes and the yellow colonies growing on the agar are examined for carotenoid content. Colonies of Flavobacterium are identified by comparing the cells to the taxonomic description provided in Bergey's Manual (1984 edition). The microorganisms are identified by their zeaxanthin content confirmed by analytical procedures such as high-performance liquid chromatography (HPLC). A method of extracting zeaxanthin involves carefully drying the cell mass, pulverizing the dried cell mass, digesting the pulverized material with an inert organic solvent, filtering the solution, and isolating the pure zeaxanthin by elution of the filtration residue with an inert organic solvent such as ethanol, acetone, or chloroform.

Lyophilized coarse-grain zeaxanthin powder

In an attempt to isolate zeaxanthin, Flavobacterium multivorum cells (ATCC 55238) were fermented to produce zeaxanthin and separated from the liquid broth by centrifugation. Acetone was added to the cells to extract a majority of the zeaxanthin into the solvent phase. The cells were then placed in a filter press to remove the cell solids and acetone was evaporated from the liquid under warm conditions (about 33 to 43 °C) and mild vacuum (Guerro-Santos and others 2005). The oily residue contained zeaxanthin in a large crystalline form and some cell residues. Water was added and the mixture was processed using a standard mechanical mixer, and forced through a Teflon filter at 15 psig pressure. The zeaxanthin and the solids that stayed on the filter were washed with hexane, a mixture of 95% hexane and 5% acetone, and then a mixture of 90% hexane and 5% acetone to remove impurities. Subsequent washings with pure acetone resulting in an acetone/zeaxanthin mixture were passed through a silica gel column. Since zeaxanthin would not cling to silica as tightly as various other impurities will, the solvent passing through the silica gel eluted the zeaxanthin from the gel. The resulting zeaxanthin, after removal of acetone, was 95% to 99% pure. Lyophilization at −70 °C and less than 100 mbar resulted in free-flowing powders at room temperature.

Zeaxanthin in viscous oily liquid

The R-R isomer produced by Flavobacterium multivorum (ATCC 55238) can be concentrated in large quantities and at low cost into a viscous oily fluid containing about 5% to 20% zeaxanthin by means of a simple solvent extraction process (Garnett and others 1998). The oily fluid may be mixed with a carrier such as vegetable oil and enclosed within a digestible capsule, comparable to a conventional capsule containing vitamin E. Alternately, a zeaxanthin fluid can be added to various types of foods such as margarine, dairy products, syrup, cookie dough, and certain types of meat preparations which are not subjected to harsh cooking. Additional purification steps can be used to produce granular zeaxanthin containing nearly pure zeaxanthin. Such processing methods may be used to create formulations such as ingestible tablets to be added to soups, salads, drinks, or other foods.

Micelles containing zeaxanthin

Zeaxanthin-containing “micelles,” which are less than 1 μm in diameter, are obtained from either the solvent extract of biomass or the oily fluid by using certain types of bile salts (Garnett and others 1998). An oily fluid may be mixed with a suitable bile salt such as the phosphate salts of glyco- or taurocholate or by use of gall bladder extracts containing mixtures of bile salts. The bile material mixed with either the solvent extract or oily mass and with salts such as sodium chloride, calcium chloride, or potassium chloride is processed in a mechanical homogenizer to produce micelles that are dried and diluted to a desired concentration using a carrier or diluent fluid such as vegetable oil. The mixture may be enclosed within a capsule or other device that will aid in swallowing and help protect the resultant micelles against degradation by stomach acid.

Microcystis aeruginosa.  High-speed counter-current chromatography has been successfully applied in the isolation and purification of zeaxanthin from the cyanobacterium Microcystis aeruginosa. Preparative high-speed counter-current chromatography with a 2-phase solvent system composed of n-hexane-ethyl acetate–ethanol–water (8:2:7:3, [v/v/v/v]) resulted in zeaxanthin of 96.2% purity in a 1-step separation (Chen and others 2005).

Chinese wolfberries and marigold flowers

Briefly, the commercial extraction of xanthophylls (oleoresin) from marigold flowers involves the following stages: ensilage, pressing, drying, hexane extraction, and saponification. The ensilage is considered critical in determining the efficiency of the overall process. Isolation of lutein and zeaxanthin esters from marigold flowers and Chinese wolfberries may involve the use of hexane as a solvent for extraction at room temperature followed by evaporation of hexane at 60 °C. The isolated oleoresin may be mixed with an alcohol to remove some of the by-product impurities.

Enzymatic treatment can enhance xanthophyll extraction from marigold flowers (Delgado-Vargas and Paredes-López 1997; Bárzana and others 2002). Petals treated with 0.1% (w/w) Econase-cep (Enzyme Development Corp., N.Y., U.S.A.) for a period of 120 h produced a significant increment in xanthophyll yield (24.7 g/kg dry weight) that compared favorably with the yield from the untreated control (11.4 g/kg dry weight) (Delgado-Vargas and Paredes-López 1997). Enhanced yields (>85%) of recovered carotenoids were also obtained from fresh flowers macerated with 0.3% (v/w flower) Viscozyme/Neutrase (Novo-Nordisk, Bagsvaerd, Denmark) and simultaneously treated with hexane (Bárzana and others 2002). However, enzymatic treatments present practical limitations owing to the high cost of commercial enzymes. Supercritical fluid extraction (SFE) is a viable alternative for many commercial separation applications to obtain carotene, lutein, and oleoresin from marigold flowers (Favati and others 1988; Naranjo-Modad and others 2000).

Lutein and zeaxanthin may also be isolated by simultaneous extraction and saponification at room temperature using tetrahydrofuran and alcoholic potassium or sodium hydroxide. To obtain pure zeaxanthin from Chinese wolfberries, zeaxanthin dipalmitate was saponified by dissolving in THF and treated with 10% KOH in methanol or ethanol (Khachik 2005). The mixture was stirred at room temperature for 1 h to complete the hydrolysis of zeaxanthin dipalmitate to zeaxanthin. The solvents were co-distilled under reduced pressure and zeaxanthin crystallized as a dark yellow-orange solid. The crystals were centrifuged and vacuum-dried at 60 °C. The yield of zeaxanthin was 16 mg from 36 mg of zeaxanthin dipalmitate (Figure 7).

Figure 7—.

Chemical structure of zeaxanthin dipalmitate (Peng and others 2005).


Supercritical fluid extraction (SFE) can extract yellow pigments from corn gluten meal, a by-product of corn starch processing (Jing 2004); optimum conditions are a temperature of 40 °C, a pressure of 20 MPa, and a time of 120 min, with 20% absolute ethyl alcohol as a cosolvent. Yields of corn pigment extracted by SFE were 2.2 times that obtained by solvent extraction. However, the pigment was found to be sensitive to sunlight and, therefore, should be stored in the dark at low temperatures.

Lipophilic carotenoids, notably (all-E)-lutein and (all-E)-zeaxanthin, and also neoxanthin, violaxanthin, β-carotene, and chlorophylls a and b from spinach and sweet corn were isolated and analyzed by high-speed counter-current chromatography (HSCCC) (Aman and others 2005a). Pretreatment with proteinase can increase the extraction efficiency of lutein, zeaxanthin, and total carotenoids from corn gluten meal (Lu and others 2005).

Orange juice

Carotenoid pigments from Brazilian Valencia orange juice were isolated by open-column chromatography (OCC) after extraction using acetone and saponification with 10% methanolic KOH (Gama and Sylos 2005). The pigments were identified as α-carotene, zeta-carotene, β-carotene, α-cryptoxanthin, β-cryptoxanthin, lutein-5,6-epoxide, violaxanthin, lutein, antheraxanthin, zeaxanthin, luteoxanthin A, luteoxanthin B, mutatoxanthin A, mutatoxanthin B, auroxanthin B, and trollichrome B.

Leaves of Physalis cups

In the leaves of Physalis, zeaxanthin occurs in the form of the palmitic acid ester physalien, from which the pigment is obtained by saponification. Physalien may also be isolated from Physalis berries (Kuhn and Wiegund 1929).

Determination of Zeaxanthin

Standards for all trans-zeaxanthin, 12′apo-zeaxanthinal, and parasiloxanthin are available from DSM (DSM Nutritional Products Ltd., Kaiseraugst, Switzerland). All-trans-zeaxanthin is also available from Fluka (Buchs, St Gallen, Switzerland).


The most common method for analysis of carotenoids is HPLC employing various detection techniques. Both normal- and reversed-phase systems, either in isocratic or gradient elution modes, are employed with reversed-phase systems being more preferred. Antioxidants are added to the mobile phase, and the column temperature is usually maintained at around 20 °C to prevent decomposition of the carotenoids during the analysis for better reproducibility of the results. A normal phase LC-atmospheric pressure chemical ionization (LC-APCI) MS-MS technique can simultaneously quantify β-carotene, α-tocopherol, β-cryptoxanthin, zeaxanthin, and lutein in biological materials (Hao and others 2005). In reversed-phase HPLC, where partition is the major chromatographic mode, the order is more or less the reverse of that encountered in normal-phase adsorption open-column chromatography. Polymeric C18 phases have excellent selectivity for structurally similar carotenoids such as the geometric isomers of β-carotene (Bushway 1985; Quackenbush and Smallidge 1986; Lesellier and others 1989; Craft and others 1990) and of lutein and zeaxanthin (Epler and others 1992). However, the total carbon load is lower in the wide-pore polymeric phases, resulting in weak retention of the carotenoids (Craft 1992). The peaks also tend to be broader and columns from different production lots are more variable than with monomeric columns. The C30 column provides an excellent resolution of photoisomerized standards of lutein, zeaxanthin, β-cryptoxanthin, α-carotene, β-carotene, and lycopene (Emenhiser and others 1995, 1996).

Temperature regulation is recommended to maintain day-to-day reproducibility. Variations in column temperature result in substantial fluctuation of the carotenoids' retention times. With a monomeric C18 column and acetonitrile–dichloromethane–methanol (70:20:10) as mobile phase, no separation of lutein and zeaxanthin, and 9-cis- and trans-β-carotene occurred at ambient (30 °C) temperature (Sander and Craft 1990). At subambient temperature (−13 °C), good separation of lutein and zeaxanthin and baseline separation of 9-cis- and trans-β-carotene were achieved. In a Vydac 201TP54 (polymeric) column with acetonitrile–methanol–dichloromethane (75:20:5) as mobile phase, optimum resolution of lutein, zeaxanthin, β-cryptoxanthin, lycopene, α-carotene, and β-carotene was achieved at 20 to 22.5 °C (Scott and Hart 1993). Also, with a Vydac 201TP column and 5% tetrahydrofuran in methanol as mobile phase, resolution of lutein and zeaxanthin and of β-carotene and lycopene was better at lower temperature (Craft and others 1992).

Cereal and cereal products A rapid procedure for the extraction and determination of carotenoids from cereals and cereal by-products involves sample saponification and extraction followed by normal-phase HPLC, allowing separation of the major carotenoids of cereals, particularly lutein and zeaxanthin (Panfili and others 2004). Among the cereals analyzed, the highest levels of carotenoids were found in corn (11.14 mg/kg dry weight), which contained β-cryptoxanthin (2.40 mg/kg dry weight), zeaxanthin (6.43 mg/kg dry weight), and α+β-carotenes (1.44 mg/kg dry weight).

Plant pigments A procedure for the simultaneous determination of lutein and zeaxanthin stereoisomers in thermally processed vegetables by HPLC with diode array detection was developed by Aman and others (2005b). (Z)-isomers of lutein and zeaxanthin were prepared by iodine-catalyzed photoisomerization. Their structures were analyzed by 1D- and 2D-LC-NMR spectroscopy, APCI-MS in the positive mode, and UV/Vis spectroscopy. Near-baseline separation was observed for (13-Z)-lutein, (13′-Z)-lutein, (all-E)-lutein, (9-Z)-lutein, (9′-Z)-lutein, (13-Z)-zeaxanthin, (all-E)-zeaxanthin, and (9-Z)-zeaxanthin.

A reversed phase-HPLC method may be used to analyze the full complement of higher plant photosynthetic pigments (cis-neoxanthin, neoxanthin, violaxanthin, anteraxanthin, lutein, zeaxanthin, cis-lutein, chlorophyll b, chlorophyll a, and α- and β-carotene) (Rivas and others 1989). The separation on a C18 column takes about 10 min, using a single high-pressure pump and 3 different mobile phases in 3 isocratic steps. The method introduces a major improvement in higher plant photosynthetic pigment analysis, resolving in all photosynthetic pigments while achieving good separation of lutein from its isomer zeaxanthin.

Positive-ion fast-atom bombardment tandem MS (FAB MS-MS) using a double-focusing mass spectrometer with linked scanning at constant B/E and high-energy collisionally activated dissociation (CAD) was used to differentiate 17 different carotenoids, including zeaxanthin (Breemen and others 1995). Carotenoids were either synthetic or isolated from plant tissues, including carrots and tomatoes. Both polar xanthophylls and nonpolar carotenes formed molecular ions during FAB ionization. Following collisionally activated dissociation, fragment ions of selected molecular ion precursors showed structural features indicative of the presence of hydroxyl groups, ring systems, ester groups, and aldehydes groups, and the extent of aliphatic polyene conjugation. It is suggested that the fragmentation patterns observed in the mass spectra may be used as a reference for structural determination of carotenoids isolated from plant and animal tissues.

Fruits, fruit juices An HPLC-diode array detector (HPLC-DAD) method was used to quantify the content of zeaxanthin dipalmitate, a major carotenoid in Fructus lycii, on a C18 column with the mobile phase consisting of acetonitrile and dichloromethane (42:58). Zeaxanthin dipalmitate was the predominant carotenoid, comprising 31% to 56% of the total carotenoids (Peng and others 2005).

An isocratic RP-HPLC method was developed for routine analysis of the main carotenoids related to the color of orange juice, using a more selective wavelength (486 nm) in which the absorption in the red-orange region of the visible spectrum is maximum. Separation was carried out using a mixture of methanol:acetonitrile:methylene chloride:water (50:30:15:5 [v/v/v/v]) to which small amounts of BHT and triethylamine were added (0.1%) as the mobile phase. Application of the method to Valencia ultrafrozen orange juices showed the major carotenoids to be lutein + zeaxanthin (36%), lutein 5, 6-epoxide (16%), antheraxanthin (14%), and β-cryptoxanthin (12%).

An HPLC-MS technique was developed for identification and determination of carotenoids and their fatty acid esters in citrus fruit juices (orange and tangerine juice concentrates) (Wingerath and others 1996). Gradient and isocratic HPLC separated as many as 38 carotenoid components in extracts from fruit juices. Structural elucidation was based on UV/vis spectroscopy, matrix-assisted laser desorption ionization (MALDI) post-source-decay (PSD) MS, and comparison with synthetic reference compounds. The xanthophylls lutein, zeaxanthin, α-cryptoxanthin, and β-cryptoxanthin were detected in extracts of saponified tangerine concentrate.

Poultry products An HPLC-DAD with a C30 phase was used for the simultaneous separation of xanthophylls in egg yolks. Peak identification was carried out by LC-(APCI) MS (Schlatterer and Breithaupt 2006).

Fish An HPLC procedure was developed to separate 2 retinol forms (retinol1 and retinol2 or dehydroretinol), their corresponding retinal forms, and carotenoid pigments commonly found in fish tissues (Guillou and others 1993). A reversed-phase C18 column was used with an isocratic solvent system (acetonitrile/dichloromethane/methanol/water/propionic acid, 71:22:4:2:1 [v/v/v/v/v]). Prior to HPLC analysis, fatty tissues were defatted with acetone/methanol (1:1 [v/v]) at −80 °C or on a silica gel column. The method gave satisfactory resolution of polar compounds (retinol and retinal forms, astaxanthin and phoenicoxanthin or zeaxanthin) and an acceptable elution time for less polar molecules (retinyl palmitate, α- and β-carotene). Levels of retinoid forms and carotenoids were determined in eyes, blood, and eggs of mature rainbow trout.


Carotenoids are so strongly colored that use of special reagents for the detection is normally unnecessary (Bolliger and Konig 1969). A number of precautions must be taken to prevent loss of pigments. These include applying the sample quickly and running the chromatogram without delay, applying the sample under subdued illumination and running the chromatogram in the dark, and using an atmosphere of nitrogen if possible. The rapid fading of carotenoids on developed thin layers may be delayed by spraying the chromatogram with a solution of liquid paraffin in light petroleum (Bolliger and Konig 1969).

Zeaxanthin can be separated from other carotenoids using a silica gel G (activated) plate and a solvent system consisting of dichloromethane:ethyl acetate (4:1 [v/v]) where lutein and zeaxanthin separate with Rf values of 0.35 and 0.24, respectively (Bolliger and Konig 1969). On silica gel G (activated) plate developed with benzene:ethyl acetate:methanol (75: 20: 5, by vol.), the Rf values of lutein and zeaxanthin are reportedly 0.57 and 0.53, respectively (Davies and others 1970a, 1970b).

Kieselguhr paper can separate carotenoids of many types and is particularly recommended as being superior to other micro-scale methods for the resolution of cis-trans isomeric xanthophylls (Liaaen-Jensen 1971). The Rf values of lutein and zeaxanthin are 0.72 and 0.59, respectively, when developed in 10% acetone in petroleum ether, and 0.91 and 0.87, respectively, in 20% acetone in petroleum ether.

NIR reflectance spectroscopy

NIR reflectance spectroscopy may be used for screening large numbers of corn samples for breeding programs aimed at producing corn hybrids with increased carotenoid concentration (Berardo and others 2004). Genotypic variation in corn carotenoid concentration was investigated in 64 genotypes, including varieties of different geographical origin, commercial hybrids, lines in selection, cornflakes, popcorn, and sweet corn. Carotenoid concentration in meals from each genotype was determined using NIR reflectance spectroscopy and HPLC. Comparison of the 2 analytical techniques revealed similar results for each; correlations of between 0.84 (lutein) and 0.94 (zeaxanthin) were identified.

Isomerization and Oxidation of Zeaxanthin

The highly unsaturated carotenoid is prone to isomerization and oxidation ( Oxidative degradation, the principal cause of extensive losses of carotenoids, depends on the availability of oxygen and is stimulated by light, enzymes, metals, and co-oxidation with lipid hydroperoxides. Carotenoids appear to have different susceptibilities to oxidation, with ζ-carotene, lutein, and violaxanthin being cited as more labile (Rodriguez-Amaya 2001). Formation of epoxides and apocarotenoids (carotenoids with shortened carbon skeleton) appears to be the initial step (Figure 8). Subsequent fragmentations yield a series of low-molecular-weight compounds similar to those produced in fatty acid oxidation. Thus, total loss of color and biologic activities are the final consequences. Heat, light, acids, and adsorption on an active surface (such as alumina) promote isomerization of trans-carotenoids, their usual configuration, to the cis forms (Figure 9).

Figure 8—.

Possible scheme for the degradation of carotenoids (

Figure 9—.

Chemical structure of zeaxanthin in the conformation: all-trans, 9-cis, 13-cis, and 15-cis.

Oxygen, especially in combination with light and heat, is highly destructive. The presence of even traces of oxygen in stored samples (even at deep-freeze temperatures) and of peroxides in solvents (for example, diethyl ether and tetrahydrofuran) or of any oxidizing agent, even in crude extracts of carotenoids, can rapidly lead to bleaching and the formation of artifacts such as epoxy carotenoids and apocarotenals (Britton 1991). Oxygen can be excluded at several steps during analysis and storage with the use of vacuum and a nitrogen or argon atmosphere. Antioxidants such as butylated hydroxytoluene, pyrogallol, and ascorbyl palmitate may also be used, especially when the analysis is prolonged. They can be added during sample disintegration or saponification or added to solvents (for example, tetrahydrofuran), standard solutions, and isolates. Exposure to light, especially direct sunlight or ultraviolet light, induces trans-cis photoisomerization and photodestruction of carotenoids. Thus, work on carotenoids must be performed under subdued light. Open columns and vessels containing carotenoids should be wrapped with aluminum foil, and thin-layer chromatography development tanks should be kept in the dark or covered with dark material or aluminum foil. Polycarbonate shields are available for fluorescent lights, which are notorious for emission of high-energy, short-wavelength radiation.

Influence of storage and thermal processing on the stability of zeaxanthin

Pure crystalline zeaxanthin degrades under the influence of atmospheric oxygen and light. Concentrated zeaxanthin stored in darkness under the inert gas argon remains stable for a month at a temperature of 40 °C, but for at least 3 y at a temperature between −3 and 5 °C. The longest storage period that has been evaluated involved zeaxanthin in a gelatin matrix, which was kept at 15 °C for 2 y and exhibited no measurable signs of degeneration (

Processing of foods may cause all-trans carotenoids to change into cis-isomers. Cis-isomers differ from trans-isomers in terms of bioavailability and antioxidant capacity (Scheiber and Carle 2005). In 1 study, all-trans zeaxanthin was incubated at 75 °C for different time intervals to study the effect of prolonged incubation at an elevated temperature on isomerization of the xanthophylls. Partial isomerization of the xanthophyll to the 13-cis conformation was found to depend on time of reaction and temperature. Formation of small amounts of the isomers 9-cis and 15-cis was shown to occur at relatively high temperatures (above 75 °C) but the rate of the formation of the 13-cis isomer was much higher (Milanowska and Gruszecki 2005). Updike and Schwartz (2003) studied the formation of cis isomers of zeaxanthin in canned corn, kale, green peas, and spinach and microwaved broccoli. Thermal processing was found to increase the cis- isomers of zeaxanthin by 17%. In zeaxanthin-rich potatoes, 70% of the initial zeaxanthin content was detected after treatment at 120 °C for 2 h (Behsnilian and others 2006). At all temperatures studied, the reduction of all-trans carotenoids was accompanied by the generation of cis-isomers, 9-cis- and 13-cis zeaxanthin in the potato matrix. However, during pasteurization of orange juice at 90 °C for 30 s, zeaxanthin content remained unaffected (Lee and Coates 2003).

A slow drying process with an irregular temperature profile (where the temperature fluctuates with the following mean values: from 22 to 34 °C in 24 h; 5 d at 44 °C; finally lowered to 5 °C in 48 h) was applied to whole pods of a Spanish pepper variety (Perez-Galvez and others 2004). The zeaxanthin losses during the first 24 h (38%) were attributed to enhanced activity of oxidative enzymes. During the temperature-holding and the cooling-down periods, only fluctuations in the zeaxanthin content were observed. Zeaxanthin losses over 80% have been reported on drying by a traditional sun-drying method for the production of paprika (Topuz and Ozdemir 2003). The zeaxanthin content of dried red pepper (pods, cut, or whole) remained stable for up to 4 mo storage at 0 °C. Increasing the storage temperature to 20 °C or the specific area of the material (powder) yielded significant zeaxanthin losses (Kim and others 2004).

Food products such as fruit juices, soy drinks, yoghurt, ice cream, biscuits, cereals and cereal bars, margarine, and soft candies with added zeaxanthin (and lutein) remain stable for periods of up to 12 mo. “Jelly bears” and extruded cereals showed losses of zeaxanthin of 25% and 78% after 9 and 12 mo, respectively (Table 4) (Koenig-Grillo 2002). Encapsulation of carotenoids with lipids, dextrins, and polyvinylpyrrolidone is 1 approach being evaluated in order to reduce carotenoid degradation in foods. Decrease in degradation during storage is reported for encapsulated carotenoids (Selim and others 2000; Basu and Del Vecchio 2001; Rodriguez-Huezo and others 2004).

Table 4—.  Stability of zeaxanthin and lutein in foods.
ProductStorage timeResults
Physical and visual stabilityRetention of lutein (%)Retention of zeaxanthin (%)
Soft drinks without juice6 moPhysically stable without pectin; not stable with pectin and 10 ppm of luteinNo chemical stability evaluated
Juice drinks6 moNo visible influence 86 92
Health eye drink3 moNo visible influence100100
YogurtProcessingNo visible influence 96100
Yogurt3 wkNo visible influence 98100
Ice creamProcessingNo visible influence 94 94
Ice cream6 moNo visible influence100100
Soy drinkProcessingNo visible influence 95 92
Soy drink6 mo., ambient temperatureTiny yellow ring, flaky sediment from juice pulp 95 92
Soy drink6 mo, 5 °CNo ring, flaky sediment from juice pulp100100
Biscuits6 moNo visible influence100 96
Extruded cereals12 moNo visible influence 21 22
Cereal bars6 moNo visible influence 99 78
Margarine6 moNo visible influence 93Not tested
Jelly bears9 moNo visible influence 70 75

The small red berry, wolfberry (Fructus barbarum L.), is one of the richest natural sources of zeaxanthin. Enhanced bioavailability of zeaxanthin in a milk-based formulation of wolfberries was obtained when homogenization of the berries was performed in hot-skimmed milk (80 °C) compared with the hot water (80 °C) and warm-skimmed milk treatment of the berries (40 °C) (Benzie and others 2006).

Chemical Synthesis of Zeaxanthin

Synthetic zeaxanthin is an orange-red crystalline powder with little or no odor. It is practically insoluble in water and ethanol, and slightly soluble in chloroform giving a clear intensive orange-red solution. It is composed of trans-zeaxanthin and minor quantities of cis-zeaxanthin, 12′-apo-zeaxanthinal, diatoxanthin, and parasiloxanthin. Nonfermentation processes suffer from several disadvantages; they typically require numerous reaction steps, and each step has a less than 100% yield, so that the final yield of zeaxanthin at the end of the multistep processing tends to be relatively poor. In addition, chemical synthesis tends to yield undesirable S-S and S-R stereoisomers of zeaxanthin, as well as various conversion products such as oxidized zeaxanthin, and zeaxanthin molecules that have lost one or more of the double bonds in the straight portion or end rings.

Zeaxanthin synthesis follows a sequence of reactions in which the final step is a double Wittig condensation of a symmetrical C10-dialdehyde as the central building block with 2 equivalents of the appropriate C15-phosphonium salt (Figure 10). The 1st step in the process is the production of an enantiopure C9-hydroxyketone either by an enantioselective catalytic hydrogenation or by a biocatalytic process combined with a chemical reduction. The enantiopure C9-hydroxyketone is then converted into the C15-phosphonium salt employed in the Wittig condensation. The C10-dialdehyde is commercially available. A similar approach is used for the synthesis of other symmetrical carotenoids, such as lycopene, β-carotene, and astaxanthin, also commercially available (Soukup and others 1996; Ernst 2002).

Figure 10—.

The Wittiz reaction used in the production of zeaxanthin (synthetic).

In the method described by Loeber and others (1971), a mixture of Wittig salt (146 mg, 0.26 mmol), 2, 7-dimethylocta-2,4,6-trienedial (16 mg, 0.1 mmol), and 1.2-epoxybutane (0.5 g) was heated for 1 h at 100 °C under nitrogen in a sealed tube, and cooled. The crude product was chromatographed on alumina (gradient elution with light petroleum, benzene, and ethyl acetate); the main red band yielded an orange solid (35 mg, 62%). Crystallization from methanol gave zeaxanthin, m.p, 211 °C; λmax (C6H6) 589, 462, and 439; Vmax, 3610, 1035, and 965 cm−l. The product did not separate from natural zeaxanthin on mixed TLC. [Kieselgel HF 254; 30% acetone in light petroleum (b.p. 60 to 80 °C)].

Zeaxanthin and its dipalmitate and physalien have been synthesized by Isler and others (1956a, 1956b). Starting from the optically pure hydroxyketone 1, an efficient synthesis of (3R, 3′R)-zeaxanthin is described by Widmer and others (1990) (Figure 11). Creemers and others (2002) synthesized pure all-E zeaxanthin as follows: TiCl3 (0.22 g, 1.4 mmol) was added to dry THF under argon; LiAlH4 (0.04 g, 1.1 mmol) was added to the suspension and the mixture was stirred for 2 h at RT. 3-Hydroxyretinal (0.21 g, 0.7 mmol) in THF was slowly added and the mixture was stirred overnight at RT. HCl (1 M) was added and the mixture was extracted twice with diethyl ether. The combined organic layers were washed with NaCl, dried with MgSO4, and concentrated in vacuo. After purification on a silica-gel column (50% diethyl ether/petroleum ether), and subsequent recrystallization (CH2Cl2/EtOH, −20 °C), pure all-E zeaxanthin (0.02 g, 0.04 mmol, 10%) was obtained.

Figure 11—.

Synthesis of (3R, 3′R)-zeaxanthin as described by Widmer and others (1990).

In vitro, the conversion of lutein to meso-zeaxanthin is readily achieved in a base-catalyzed reaction (Bone and others 1993), and this is the basis of an industrial process for its synthesis and use in poultry feeds. The conversion hypothesis is supported by the distribution of the individual carotenoids in the retina (Bone and others 2007).

Acceptable Daily Intake of Zeaxanthin

The safety of zeaxanthin in supplements has been confirmed by the New Dietary Ingredient Notification for dietary 3R, 3R′ zeaxanthin. The notification was filed by Roche Vitamins and was reviewed by FDA without objection in 2001 (Docket Nr 95S-0316: 75-Day Premarket Notifications for New Dietary Ingredients, Report Nr 96). In addition, the Joint Food and Agriculture Organization/World Health Organization (FAO/WHO) Expert Committee on Food Additives (JECFA) recently completed a toxicological evaluation of zeaxanthin. JECFA established a group Acceptable Daily Intake (ADI) for lutein and zeaxanthin of 0 to 2 mg/kg body weight for zeaxanthin (JECFA, Summary and Conclusions, 63rd meeting, Geneva, June 2006). There are also direct animal data demonstrating retinal protective effects of dietary zeaxanthin (Thomson and others 2002).

Functional Properties of Zeaxanthin

The intake of a carotenoid-rich diet is epidemiologically related to a lower risk of some types of cancer and AMD (Perez-Galvez and others 2003). Zeaxanthin is used in the prevention of human macular degeneration and also for poultry pigmentation (Berry and others 2003). For infants, human milk is the main source of lutein and zeaxanthin until weaning occurs. The carotenoids appear to be important as protective factors in the retinal pigment epithelium of the newborn infants (Jewell and others 2001, 2004). The evaluation of zeaxanthin in the cosmetic area, especially in solar protection, is under way owing to its natural function as a protectant of cells from photosensitization.

Nutraceutical value of zeaxanthin

Age-related macular degeneration (AMD) AMD, the leading cause of irreversible blindness in adults, is the result of degenerative changes that occur in the central region of the retina and the macula, eventually leading to loss of vision. “Dry” macular degeneration is caused by a thinning of the macula's layers, and vision loss typically is gradual. However, tiny, fragile blood vessels can develop underneath the macula. “Wet” macular degeneration can result when these blood vessels hemorrhage, and blood and other fluid can further destroy macular tissue, even causing scarring. In this case, vision loss can be rapid, over months or even weeks. Lutein and zeaxanthin may play a major role in the prevention of AMD. They are not metabolized to vitamin A, and they accumulate in the macular region of the human retina, where they prevent damage by absorbing high-energy blue light through their antioxidant properties (Handelman and others 1988; Britton 1995; Snodderly 1995). It is suggested that people supplement their diets with 4 mg of lutein/zeaxanthin daily. Dietary studies have confirmed the association between frequent consumption of spinach or collard greens, particularly good sources of lutein and zeaxanthin, and lower AMD risk ( Foods provide other phytochemicals that might act in a synergistic way to help the absorption and utilization of the nutrient or to aid in the way it protects the body from damage.

Zeaxanthin and lutein are specifically accumulated in the macular region of the retina where they bind to the retinal protein, tubulin (Handelman and others 1988, 1991). Zeaxanthin is specifically concentrated at the macula, whereas lutein is distributed throughout the retina. Within the central macula, zeaxanthin is the dominant component, up to 75% of the total, whereas in the peripheral retina, lutein predominates, usually being 67% or greater. Khachik and others (2002) found the ratio of lutein to dietary 3R, 3R′–zeaxanthin in the lens to be approximately 1.5:1, which suggests a preferential uptake for dietary 3R, 3R′–zeaxanthin from the diet and plasma to the lens. The human retina accumulates lutein and zeaxanthin together with nondietary metabolites of lutein and zeaxanthin including meso-zeaxanthin [(3R, 3′S)-β, β-carotene-3, 3′diol] and 3′-oxolutein [(3-hydroxy-3′-oxo-β-ɛ-carotene)] (Bernstein and others 2001). A significant fraction of lutein in the central retina is converted into meso-zeaxanthin. In the center of the retina, the ratio of meso-zeaxanthin to lutein is the highest and approaches 1:1. Data from autopsy eyes indicate that the (lutein + meso-zeaxanthin):zeaxanthin ratio varies throughout the retina from 2:1 in the center, where most of the conversion occurs, to 3:1 in the periphery (Bone and others 1997). Thus there may be an advantage in providing meso-zeaxanthin in supplements at the expense of lutein if the goal is to raise the overall zeaxanthin (Bone and others 2007). Lycium Chinese Mill, also known as Chinese boxthorn used in traditional Chinese medicine for the treatment of a number of disorders, including visual problems, has zeaxanthin in its fatty acid ester form as the principal carotenoid.

Cataract Zeaxanthin and lutein may also be protective against age-related increase in lens density and cataract formation. Cataracts are the leading cause of impaired vision, with a large percentage of the geriatric population exhibiting signs of the lesion. Cataracts are developmental or degenerative opacities of the lens that result in a gradual, painless loss of vision. Studies examining lutein and zeaxanthin levels in extracted cataractous lenses have found up to 3-fold higher levels in the newer epithelial tissue of the lens than in the older inner cortex portion. The epithelial cortex layer comprises 50% of the tissue, yet it has been found to contain 74% of the total lens lutein and zeaxanthin, supporting the hypothesis that these nutrients are protective against the oxidative damage causing cataract formation (Yeum and others 1999).

Conjugated double bonds are particularly effective at absorbing harmful UV rays and also quenching singlet oxygen that produces reactive oxygen species (ROS). Zeaxanthin, which is fully conjugated, may offer better protection than lutein against oxidative damage of the lipid matrix caused by the blue and near-ultraviolet light radiation in macular membranes. A protective effect of lutein and zeaxanthin against the oxidative damage of egg yolk lecithin liposomal membranes induced by exposure to UV radiation and incubation with 2,2′-azobis(2-methypropionamidine)dihydrochloride, a water-soluble peroxidation initiator, is shown by Sujak and others (1999). However, both lutein and zeaxanthin were found to protect lipid membranes against free radical attack with almost the same efficacy. As a free-radical scavenger, β-carotene and zeaxanthin, with the same chromophore, are less effective than lycopene. Lycopene is considered to be the most effective free radical scavenger owing to its 11 conjugated coplanar double bonds.

Cancer Zeaxanthin has cancer-preventive properties. Spontaneous liver carcinogenesis in C3H/He male mice was suppressed when fed for 40 wk with zeaxanthin at a concentration of 0.005% and mixed as an emulsion in drinking water (Nishino and others 1999).

Inhibition of LDL oxidation The carotenoid plays an important role in the inhibition of macrophage-mediated LDL oxidation (Keri and others 1997). Zeaxanthin when incubated with human monocyte-macrophages for 24 h with human low-density lipoprotein (LDL) was found to inhibit LDL oxidation in a concentration-dependent manner, suggesting that zeaxanthin might help in slowing atherosclerosis progression.

Pigmentation of poultry and fish

Zeaxanthin is preferred over other carotenoids for enhancing pigmentation in poultry and fish due to its potency to provide a true color and ability to deposit evenly in the flesh and eggs (Orndorff and others 1994). Zeaxanthin is at least 1.5 times as potent as lutein. When administered to poultry in high doses, carotenoid or carotenoid-containing compounds such as canthaxanthin, alfalfa, and cayenne pepper cause abnormal red or purple colors in the flesh and color striations in yolks. High doses of lutein in the feed have been shown to impart a greenish hue to poultry flesh and egg yolks. Beta-carotene is not deposited well in the flesh of poultry, and canthaxanthin apparently deposits in the iris of the eye. Zeaxanthin, in contrast, imparts a yellow-red color even at high doses and deposits well in poultry flesh, based on high zeaxanthin-containing corn feed studies. Labeled zeaxanthin fed to laying hens undergoes conversion to (6S,6′S)-ɛ, ɛ-carotene-3,3′ dione and the intermediate (3R,3′S)-3-hydroxy-β, ɛ-carotene-3′-one, both of which appear in the egg yolks (Schiedt and others 1981). Some fish and crustaceans, such as shrimp, goldfish, and carp, apparently can convert zeaxanthin into the red-colored pigment astaxanthin, suggesting that feeding of zeaxanthin to such fish and crustaceans will enhance desirable red coloration. The pigment-containing biomass can be used to color foodstuffs without the necessity for isolating pure zeaxanthin (Gierhart 1994). For poultry products, there are problems with stability and biological availability, particularly with xanthophylls from marigold and alfalfa. Most marigold products must be solvent-extracted and saponified, and may require the inclusion of antioxidants in the extraction process (Gierhart 1994).

Absorption of Zeaxanthin

Humans cannot synthesize carotenoids and, therefore, must rely on dietary sources to provide sufficient levels. The human body absorbs zeaxanthin in the same way as other carotenoids and fat-soluble vitamins. There is no difference between synthetic and natural zeaxanthin in terms of the way they are processed by the human system. The average concentration (ng/g) of zeaxanthin in human tissues and skin is reported to be liver, 591; lung, 90; breast, 14; prostate, 35; colon, 32; and skin, 6 (Khachik and others 1998; Scholz and others 2000). Average carotenoid concentration (ng/tissue) in human ciliary and RPE/choroid is 2.54 and 4.85, respectively (Bernstein and others 2001).

The biological availability of zeaxanthin depends not only on the composition of the matrix containing it, but also on other foods consumed at around the same time. The major factors limiting the bioavailability include molecular structure, interactions of xanthophylls with other nutrients (mainly dietary fat), the manner in which the food is processed, the physical form, speciation, molecular linkage of the carotenoid in the plant tissue, the nutritional status and physiological state of the subject, and genetic background and the physical disposition of xanthophylls in the food matrix (Yeum and Russell 2002). Carotenoids are best absorbed in the presence of fat, but as little as 3 to 5 g in a meal appear to ensure carotenoid absorption (van Het Hof and others 2000a). Zeaxanthin in egg yolks might be highly bioavailable because of its association with the lipid matrix of the egg yolk (Handelman and others 1999). The carotenoids in egg yolk are in a digestible lipid matrix consisting of cholesterol (200 mg/yolk), phospholipids (1 g/yolk), and triacylglycerols (4 g/yolk) (Human Nutrition Information Service 1989). Such a lipid matrix may be optimal for carotenoid absorption from the diet. Only about 5% of the carotenoids in whole, raw vegetables are absorbed by the intestine, whereas 50% or more of the carotenoid is absorbed from micellar solutions. Thus, the physical form in which the carotenoid is presented to intestinal mucosal cells is of crucial importance (Olson 1994).

The 1st step in carotenoid bioavailability is release from the food matrix. Food processing activities such as thermal processing, mincing, or liquefying result in changes to carotenoid chemistry, probably through isomerization or oxidation reactions (Kopsell and Kopsell 2006). However, freezing or low-temperature storage generally preserves carotenoid concentration by reducing potential enzymatic oxidation. Processing activities usually increase bioavailability through increased release of bound carotenoids from the food matrix; however, thermal degradation in carotenoid chemistry might adversely affect bioavailability in some food crops.

Carotenoids are lipid-soluble molecules that follow the absorption pathway of dietary fat (Figure 12). Early in the digestive process, carotenoids are partially released from the food matrix by mastication, gastric action, and digestive enzymes (Deming and Erdman 1999). Carotenoids released from the food matrix migrate and solubilize into lipid globules of varying sizes in the stomach, where they are eventually transformed into smaller lipid emulsion particles by normal digestive motility. Solubilization of individual carotenoid molecules into lipid emulsions is thought to be a selective process related to the specific polarity of each molecule. Nonpolar carotenes most likely migrate to the triacylglycerol-rich core of the particle, while the more polar xanthophylls orient at the surface monolayer along with proteins, phospholipids, and partially ionized fatty acids. Borel and others (1996) reported this selective orientation of carotenoids in phospholipid-triacylglycerol droplets using an in vitro biological emulsion model. Beta-carotene, a nonpolar carotene, was shown to solubilize in the core of the droplets, while zeaxanthin, a more polar xanthophyll, accumulated at the surface. Selective orientation of carotenoids in membranes, micelles, and lipoproteins is most likely similar to that of other lipid molecules and is based upon the polarity, length, and structure of the molecule (Deming and Erdman 1999). Carotenoids solubilized in lipid emulsion particles are transported from the stomach to the duodenum of the small intestine (Deming and Erdman 1999). Dietary fat in the duodenum triggers the release of bile acids from the gall bladder and regulates levels of pancreatic lipase. Bile acids aid in the reduction of lipid particle size and stabilization into mixed micelles. Dietary fat increases the synthesis and secretion of pancreatic cholesteryl esterase (CEL) and chylomicrons. CEL is capable of hydrolyzing triacylglycerols, phospholipids, cholesteryl esters, and vitamin A and E esters in the presence of bile salts. Carotenoids are solubilized into mixed micelles along with dietary triacylglycerols, their hydrolysis products, phospholipids, cholesterol esters, and bile acids (Deming and Erdman 1999). Studies on the capacity of CEL to hydrolyze zeaxanthin esters incorporated into micelles showed mixed micelles containing zeaxanthin esters extracted from wolfberry and incubated with CEL (1000 unit/L) decreased by 84% after 4 h (Chitchumroonchokchai and Failla 2006). The decrease in zeaxanthin diesters was majorly associated with the appearance of free zeaxanthin in the micelles. In contrast, free zeaxanthin was not detected in the medium without CEL.

Figure 12—.

Pathway of carotenoid absorption and metabolism: C = carotene; X = xanthophylls; LPL = lipoprotein lipase; HDL = high-density lipoprotein; VLDL = very low-density lipoprotein; LDL=low-density lipoprotein.

Uptake of carotenoids into the intestinal mucosa occurs by passive diffusion (Deming and Erdman 1999). The process requires solubilization of the mixed micelle in the unstirred water layer surrounding the microvillus cell membrane of the enterocyte. The mixed micelles collide and diffuse into the membrane, releasing carotenoids and other lipid components into the cytosol of the cell. Parker (1996) suggested a concentration gradient between the micelle and the cell membrane to most likely determine the rate of diffusion. High doses of carotenoid may saturate uptake leading to insufficient removal from the plasma membrane, thereby reducing this concentration gradient. Carotenoid uptake into the enterocyte does not ensure its metabolism or absorption in the body. Carotenoids in the enterocyte may be lost in the lumen of the gastrointestinal tract due to normal physiological turnover of the mucosal cells (Erdman and others 1993). Selective surface orientation of more polar xanthophylls into micelles suggests that their uptake into the enterocyte and incorporation into chylomicrons may occur before that of nonpolar carotenes. Gartner and others (1996) reported the preferential uptake of lutein and zeaxanthin from the intestinal lumen into chylomicrons in humans, even in the presence of high amounts of β-carotene. Interestingly, while most xanthophylls are present in fruits as esters (Khachik and others 1991), only free xanthophylls have been found in the chylomicrons and serum of humans (Wingerath and others 1995), suggesting a requirement for hydrolysis of carotenoid esters prior to uptake and incorporation into chylomicrons. Several human studies show that free xanthophylls are present in the triglyceride-rich fraction of plasma after ingestion of a meal enriched with esterified zeaxanthin (Wingerath and others 1995; Breithaupt and others 2004) and cryptoxanthin (Breithaupt and others 2003). Wingerath and others (1995) also reported an increase in the concentration of free cryptoxanthin, zeaxanthin, and lutein in chylomicrons and sera of human subjects after they ingested tangerine juice concentrate that was rich in the esterified forms of the xanthophylls. These observations suggest that carotenoid esters are hydrolyzed before incorporation into nascent chylomicrons synthesized in the absorptive intestinal epithelial cells.

Chylomicrons are secreted from the enterocyte into the lymphatic circulation for transport to the liver. Prior to hepatic uptake, chylomicrons in the bloodstream are rapidly degraded by lipoprotein lipase associated with tissue endothelium and transformed into chlylomicron remnants. During this process, some carotenoids may be taken up by extrahepatic tissues (Deming and Erdman 1999). However, most chylomicron remnants deliver carotenoids to the liver where they are stored or resecreted into the bloodstream in very low density lipoproteins (VLDL) (Johnson and Russell 1992). Circulating VLDLs are subsequently delipidated to low-density lipoproteins (LDL). Similar to other nonpolar lipids, carotenes such as β-carotene and lycopene are thought to migrate to the hydrophobic core of lipoproteins, while the more polar xanthophylls reside closer to the surface where the likelihood of exchange among lipoproteins is enhanced (Romanchik and others 1995). In the fasting state, LDLs are the main carriers of nonpolar carotenoids, β-carotene, α-carotene, and lycopene, in human serum (Pateau and others 1998). The more polar xanthophylls are evenly distributed between high-density lipoproteins (HDL) and LDL, and to a lesser extent VLDL. Carotenoids released from lipoproteins, especially LDL, are taken up by extrahepatic tissues.

Red pepper (Capsicum annuum L.) and its dietary products containing a variety of carotenoids may contribute to the carotenoid pattern of human blood and tissues. The availability of carotenoids from paprika oleoresin, including zeaxanthin, β-cryptoxanthin, β-carotene, and the paprika-specific oxocarotenoids, capsanthin, and capsorubin, was assessed in human volunteers (Perez-Galvez and others 2003). At different time points, the carotenoid pattern in the chylomicron fraction was analyzed to evaluate carotenoid absorption. Of the major carotenoids present in paprika oleoresin, only zeaxanthin, β-cryptoxanthin, and β-carotene were detectable in considerable amounts. Although the xanthophylls were mainly present as mono- or diesters, only free zeaxanthin and β-cryptoxanthin were found in the human samples. Although the bioavailability of the pepper-specific carotenoids, capsanthin, and capsorubin from paprika oleoresin is very low, the oleoresin is a suitable source of the provitamin A-carotenoids, β-carotene, and β-cryptoxanthin, and the macular pigment zeaxanthin.

Assessment of Bioavailability of Zeaxanthin

In vivo methods

The in vivo bioavailability studies imply the consumption of a certain dose of a nutrient and changes of its concentration (measured by standard analytical procedures) in the blood plasma compared with time (such as postprandial period, or the time after the meal) (Parada and Aguilera 2007). Three parameters are derived from the kinetics: the area under the curve (AUC), the maximal plasma concentration (Cmax), and the time to reach Cmax, tmax. AUC is a measure of the absorption intensity, whereas Cmax and tmax give an idea of the rate of absorption (Heacock and others 2004; Manach and others 2005). Another method to assess bioavailability is to measure the plasma concentration of a nutrient through an extended period (days, weeks) of constant consumption of a specific food (van het Hof and others 1999; Richelle and others 2002). Relevant parameters in this case are Csat (value at which the concentration remains constant in the time) and tsat (time at which Csat is attained).

A single-blinded, placebo-controlled, human intervention trial was used to provide data on the effect of dietary supplementation with whole wolfberries on fasting plasma zeaxanthin concentration (Cheng and others 2005). A total of 14 human subjects took 15 g/d wolfberry (estimated to contain almost 3 mg zeaxanthin) for 28 d. Age- and sex-matched controls (n= 13) took no wolfberry in their diet. On supplementation, plasma zeaxanthin concentration increased 2.5-fold, indicating zeaxanthin in whole wolfberries to be bioavailable and that intake of a modest daily amount could markedly increase fasting plasma zeaxanthin levels.

The influence of saponification on the deposition in egg yolks of carotenoids from the oleoresin of red pepper (Capsicum annuum) was studied by Hamilton and others (1990). Four replicates of 10 hens depleted of carotenoid stores were fed for 3 wk with a basal diet of white corn and soybean meal, amended with (1) saponified or (2) nonsaponified oleoresin. Trans-lutein, trans-zeaxanthin, and trans-capsanthin accounted for >85% of the total carotenoids deposited in the yolks. From the ratios of total carotenoid concentration in the yolks to those in the feed, the carotenoids from (1) were deposited twice as efficiently as those from (2).

The main drawbacks of in vivo data are the variability in physiological state of individuals and the possible interaction of the nutrient with other components in the diet (Boileau and others 1999). Whole plasma pharmacokinetics may not be the most practical way to measure carotenoid status as plasma concentrations are not only a measure of the absorption of carotenoids from the diet, but also a measure of exchange from tissue storage, bioconversion, and excretion (Castenmiller and West 1998). Also, ethical restrictions to abide by severe protocols when humans and/or animals are used in biological research severely limit these types of studies (van het Hof and others 2000b).

In vitro methods/gastrointestinal models (GIMs)

In vitro methods are being extensively used since they are rapid, safe, and do not have the ethical restrictions of in vivo methods. They are used as a suitable alternative to in vivo assays to determine bioavailability. In vitro methods simulate either the digestion and absorption processes (for bioavailability) or only the digestion process (for bioaccessibility), and the response measured is the concentration of a nutrient in the final extract. The digestion process is simulated under controlled conditions using commercial digestive enzymes (for example, pepsin, pancreatin, and so on) while the final absorption process is commonly assessed using “Caco-2 cells” cultures. “Caco-2 cells” is the short name for polarized human colon carcinoma cells line (Verwei and others 2005). Methods that simulate under laboratory conditions the gastrointestinal digestion process are known as gastrointestinal models (GIMs) (Parada and Aguilera 2007).

The in vitro digestion/Caco-2 cell culture procedure is a rapid and cost-effective model for screening the bioavailability of carotenoids from plant foods (fresh spinach, fresh carrots, and tomato paste) (Figure 13). Chitchumroonchokchai and others (2004) studied lutein bioavailability by simulating the gastric and small intestinal phases of digestion, followed by a Caco-2 cell assay. According to Chitchumroonchokchai and Failla (2006), ingested xanthophyll esters are processed in a manner similar to dietary cholesteryl esters in the lumen of the small intestine. With differentiated cultures of Caco-2 intestinal cells and CEL-null mice, CEL enhanced intestinal uptake and transport of cholesterol by cleaving cholesteryl esters both before and after their partitioning in micelles to increase the extracellular concentration of free cholesterol. In contrast, CEL does not affect apical uptake of either free cholesterol or free zeaxanthin. Once taken up by the enterocyte, the metabolism of cholesterol and the xanthophylls differs. Cholesterol is esterified by acyl-CoA:cholesterol acyltransferase and incorporated into nascent chylomicrons for secretion into lymph. In contrast, the concentration of free zeaxanthin remaining relatively constant in Caco-2 cells suggests that enterocytes do not esterify xanthophylls before their transfer to nascent chylomicrons. The cells preferentially took up free zeaxanthin from micelles and the extent of transport (4.5%) in chylomicrons was similar to that reported for lutein in Caco-2 cells (Chitchumroonchokchai and others 2004). This low amount is consistent with the estimated absorption of 3.3% of 5 mg zeaxanthin administered to human subjects. Studies are now required to determine whether, as expected, zeaxanthin uptake occurs in a manner similar to that of lutein and to examine possible interactions between zeaxanthin and the other dietary carotenoids that affect their absorption.

Figure 13—.

In vitro method to determine bioavailability involving a digestion absorption step using a Caco-2 cell culture (Parada and Aguilera 2007).

Commercial Zeaxanthin

Naturally occurring carotenoids are of commercial interest as coloring agents for foods, pharmaceuticals, cosmetics, and animal feeds. Owing to their antioxidant properties, most of them have been proposed in the prevention of chronic diseases (Rao and Agarwal 1999). Although lutein is almost always accompanied by zeaxanthin, the reverse is not true. Pure commercial dietary 3R, 3R′ zeaxanthin contains no lutein.

As a supplement, lutein and zeaxanthin are available in either the free or esterified forms, which appear to have comparable bioavailability (Bowen and others 2002). Very few commercial carotenoid mixtures contain more than extremely small or trace quantities of zeaxanthin (Garnett and others 1998). A commercial carotenoid mixture that lists zeaxanthin as one of the carotenoids contained in their carotenoid mixtures is the “β-carotene formula preparation,” sold by General Nutrition Corp. The great majority of carotenoids in the carotenoid mixtures are β-carotene and vitamin A. Xangold 10% beadlets, a dark, reddish-brown tablet grade powder containing natural mixed carotenoid esters isolated from marigold flowers, comprise mainly lutein esters with very small amounts of zeaxanthin esters. Although commercial lutein/zeaxanthin supplements often contain significantly more lutein than zeaxanthin, new products are being developed with higher amounts of zeaxanthin. A new dietary supplement, with zeaxanthin and lutein in the same 2:1 ratio as found in a healthy macula, is now available as “EyePromise Restore” (ZeaVision, LLC, Saint Louis, Mo., U.S.A.). The supplement contains a patented dose of zeaxanthin and lutein, the carotenoids that make up the pigment in the macula (


Zeaxanthin is effective against AMD, cataract, and LDL oxidation, thus emphasizing its use in nutraceutical formulations. Bile acids may play a role in the cellular uptake of the carotenoid. The xanthophyll ester is hydrolyzed before absorption, and dietary fat greatly stimulates the absorptive process. Zeaxanthin is preferred over other carotenoids for enhancing pigmentation in poultry and fish due to its ability to deposit evenly in the flesh and eggs. Flavobacterium sp., which synthesizes zeaxanthin as the major pigment, has been widely studied for the production of the carotenoid. In the aerobic fermentation for the production of zeaxanthin, oxygen supply to microorganisms is one of the variables that invariably influence the yield. Although many fermentation studies have been conducted for the production of zeaxanthin, very little information is available on pigment regulation by nutritional factors.