RESEARCH ARTICLES
`
`Abstract
`
`The Bioavailability of Astaxanthin Is Dependent on
`Both the Source and the Isomeric Variants of the
`Molecule
`
`Myriam MIMOUN-BENARROCH1, Cindy HUGOT1, Larbi RHAZI1, Claude-Narcisse NIAMBA1, Flore
`DEPEINT1*
`1 Institut Polytechnique LaSalle Beauvais-ESITPA (UniLaSalle), 19 rue Pierre Waguet, 60026 Beauvais
`cedex, France
`* Corresponding author: flore.depeint@unilasalle.fr
`Bulletin UASVM Food Science and Technology 73(2)/2016
`ISSN-L 2344-2344; Print ISSN 2344-2344; Electronic ISSN 2344-5300
`DOI: 10.15835/buasvmcn-fst:12350
`Astaxanthin is a marine carotenoid that has a number of potential health benefits, including a very strong
`antioxidant potential. Present in the flesh of salmonids and shellfish, its natural sources currently on the market
`for food supplements come from the algae Haematococcus pluvialis and krill. However other natural sources can
`be found and may be of interest.Cellular uptake studies were performed on Caco-2/TC7 colonic cells. The cells
`were cultured on a semi-permeable membrane to create a polarized and functional epithelium, representative of
`the intestinal barrier. Four sources of astaxanthin were selected and compared; synthetic, natural extracts from
`bacteria, algae or yeast. Astaxanthin was incorporated at a concentration of 5µM into mixed micelles and applied
`to cultured cellsand concentration of astaxanthin measured by HPLC in both apical and basolateral compartments.
`Small variations in bioavailability were observed at 3 hours. After 6 hours, only the algae source of astaxanthin
`was still present in the apical compartment as the esterified form. Structure-activity relationships are further
`discussed. Animal experiments usingyeast and algae sources in different types of matrices confirm the role of
`source and formulation in the bioavailability potential of astaxanthin.
`Keywords: astaxanthin,bioavailability, Caco-2 cells, carotenoids, lipid metabolism
`proved the most effective, followed by zeaxanthin
`and beta-carotene (Tinkler et al., 1994). Astaxan-
`Astaxanthin belongs to the family of carot-
`thin also neutralizes free radicals. A study shows
`enoids and has powerful antioxidant properties.
`that it is 50 times more effective in preventing
`Increasing interest has been targeted toward this
`peroxidation of fatty acids that beta-carotene or
`molecule and its biological benefits in recent dec-
`zeaxanthin(Terao, 1989). In most aquatic animals
`ades. Several studies compared the antioxidant ac-
`in which it can be found, astaxanthin has several
`tivity of astaxanthin with other carotenoids. One
`essential biological functions, including protection
`study found that astaxanthin neutralized twice as
`against the oxidation of polyunsaturated essential
`efficiently singlet oxygen as beta-carotene (and
`fatty acids protect against the effects of UV light,
`almost 80 times more effective than vitamin E) in
`immune response modulation, pigmentation,
`chemical solution (Mascio et al. , 1991). Lycopene
`communication, reproductive behaviour and the
`by comparison was 30% more efficient than astax-
`improvement of reproduction (Lorenz &Cysewski,
`anthin. Similar results were observed by research-
`2000). Astaxanthin cannot be synthesized by most an-
`ers working on in vitro system of human blood
`cells treated with various carotenoids and then
`imals and must be acquired from the diet. Although
`exposed to singlet oxygen. Again, lycopene has
`
`
`
`INTRODUCTION
`
`RIMFROST EXHIBIT 1099 page 0001
`
`

`

`62
`
`mammals and most fish are unable to convert other
`dietary carotenoids to astaxanthin, crustaceans
`(such as shrimp) and some fish species (koi) have
`a limited ability to convert closely related dietary
`carotenoids into astaxanthin, although most astax-
`anthin recovered comes from their diet. Mammals
`do not have the capacity to synthesize astaxanthin
`and unlike beta-carotene astaxanthin has no pro-
`vitamin A activity in these animals.
`Astaxanthin is naturally present in the human
`diet with seafood such as krill, shrimp, lobster, cod,
`mackerel, salmon or other coloured fish. In wild
`salmon, concentrations of astaxanthin can reach
`40 mg/kg while farmed salmon can only reach up
`to 8 mg/kg(Ambatiet al., 2014). The daily intake of
`2-4 mg astaxanthin, recommended for physiologi-
`cal effects, thus corresponds to the absorption of
`100 g of wild salmon or 500 g of farmed salmon.
`The main non-animal natural sources of astaxan-
`thin are the microalgae Haematococcus pluvialis
`and yeast Phaffia rhodozyma. Only algae astaxan-
`thin is currently marketed as dietary supplement
`for humans; Phaffia rhodozyma yeast extracts on
`the other hand are currently only used for animal
`feed (fish, eggs). The synthetic source is also wide-
`ly used in animal feed, while production by the
`bacterium Paracoccuscarotinifaciens is more anec-
`dotal. These four sources present specific geomet-
`ric and optical isomers detailed in Tab. 1.The asta-
`xanthin profile being identical in fish flesh to that
`present in their diet (Storebakken et al., 1985), the
`different forms used in aquaculture therefore find
`themselves indirectly in the human diet.
`The various stages of transportation, diges-
`tion, absorption and transport in the plasma of
`dietary carotenoids were examined in mammals
`(Furr& Clark, 1997)but also because these com-
`pounds have been associated with reducing risks
`of certain cancers and chronic diseases. Full un-
`derstanding of carotenoid metabolism is compli-
`cated by a number of factors: variations in phys-
`iochemical properties among carotenoids; altered
`carotenoid utilization as a result of the normal
`vicissitudes of lipid absorption and transport; di-
`vergence in metabolic fate within the intestinal
`enterocyte (especially carotenoid cleavage to reti-
`noids. In plasma, the non-polar carotenoids such
`as beta-carotene, alpha-carotene or lycopene are
`usually transported by very low density lipopro-
`teins (VLDL) and low density lipoproteins (LDL)
`and the polar carotenoids such as zeaxanthin,
`
`MATERIALS AND METHODS
`
`MIMOUN-BENARROCH et al.
`lutein or astaxanthin are more likely to be trans-
`ported by LDL and high density lipoprotein (HDL).
`Similarly, a limited number of clinical studies have
`investigated the bioavailability of astaxanthin
`from algae (Mercke-Odeberget al., 2003; Okada,
`Ishikura, &Maoka, 2009) or a synthetic esterified
`form (Coral-Hinostrozaet al., 2004; Østerlie et al.,
`2000)pharmacokinetics, and distribution of astax-
`anthin E/Z and R/S isomers in plasma and lipo-
`protein fractions were studied in 3 middle-aged
`male volunteers (37-43 years but none so far on
`fermentative sources such as yeast or bacteria.
`To our knowledge, no report comparing the
`different sources of astaxanthin has been pub-
`lished. This study therefore aims to compare on
`a cellular epithelial transport model four sources
`of astaxanthin used in animal feed, some of which
`are used or intended for human consumption.
`All reagents were purchased from Sigma Al-
`drich (France), except for the different astaxan-
`thin extracts which were generously donated by
`Ajinomoto Foods Europe, Algatechnologies ltd and
`ACS Dobfar spa for the yeast, algae and bacterial
`sources, respectively. We followed an existing pro-
`tocol (Mimoun-Benarroch et al., 2011) to mimic
`the intestinal absorption of lipophilic molecules
`solubilized in the form of bile salt micelles. This
`formulation, based on the composition of post-
`prandial duodenal contents in humans (Armand
`et al., 1996), appears to be the closest to physi-
`ological conditions. In brief, cholesterol (0.1 mM),
`phosphatidyl choline (0.5 mM), lyso-phosphatidyl
`choline (1.5 mM), monoolein (0.03 mM), sodium
`oleate (0.5 mM) and astaxanthin (5 µM) were
`mixed in a methanol:chloroform solution (2/1;
`v/v) and evaporated under a stream of nitrogen.
`Sodium taurocholate (5 mM) was then diluted in
`half a volume of culture medium (DMEM) with-
`out serum and without phenol red and added to
`the dried lipid residue and vigorously mixed by
`sonication at 25 W for 3 min. When the medium
`is translucent, the second volume of culture me-
`dium is added and the solution stirred overnight
`to allow the micelles to stabilize. The solution is
`filtered (0.2 µm) before treatment to retain only
`the uniform size of micelles.
`Media and cell culture solutions were pur-
`chased from Life Technologies (France). Caco2/
`TC7 cellswere a generous gift from Dr Monique
`
`RIMFROST EXHIBIT 1099 page 0002
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`The Bioavailability of Astaxanthin Is Dependent on Both the Source and the Isomeric Variants of the Molecule
`Tab. 1. Sources of astaxanthin (AX) and their structural variations
`
`63
`
`AX
`(ppm)
`
`Optical isomers (%)
`
`3R,3’R 3R,3’S
`
`3S,3’S
`
`Geometrical
`isomers (%)
`All-
`trans
`
`Cis
`
`Derivation (%)
`Free Mono-
`ester Diester
`
`Source
`
`farmed salmon
`
`wild salmon
`
`krill
`
`Phaffia rhodozyma
`
`Paracoccuscarotinifaciens
`
`Haematococcus pluvialis
`
`Synthetic
`
`2.5-8
`
`5-30
`
`25
`
`12-17
`
`50
`
`2-6
`
`120
`5000-
`10000
`20000
`10000-
`40000
`100000
`
`9-55
`
`7-21
`
`98
`
`-
`
`4
`
`-
`
`-
`
`8
`
`25
`
`50
`
`25
`
`78-85
`
`38-70
`
`2
`
`100
`
`88
`
`75
`
`25
`
`70-90
`
`10-30
`
`95
`
`70
`
`5
`
`30
`
`4-5
`
`100
`
`100
`
`5
`
`31-35
`
`61-64
`
`-
`
`-
`
`85
`
`-
`
`-
`
`15
`
`-
`
`25
`
`65-75
`
`25-35
`
`100
`
`-
`
`incubation by collection of the different media
`(apical, basolateral). Finally the cellular layer was
`washed with PBS before being scraped from the
`insert and collected in 500 µl PBS. Samples were
`kept frozen (-80 °C) until analysis.
`The extraction protocol is adapted from
`Mercke-Odeberg et al(2003). Briefly, 6 volumes of
`hexane and 6 volumes of acetone were added, the
`sample vortexed for 1 min before centrifugation
`for 5 min at 2500 rpm, room temperature. The
`organic phase was transferred to a new tube and
`evaporated under nitrogen flux. The astaxanthin
`pellet was recovered in 200 μL acetone for HPLC
`analysis. In the case of tests on the algae source
`of astaxanthin, the extraction residue was taken
`up in 1.5 ml of acetone supplemented with 1.5 mL
`of cholesterol esterase solution (Sigma) in 50 mM
`Tris HCl pH 7 (4 U / ml) and incubated for 2 h at
`37 °C, then 0.5 g of sodium decahydrate and 1 mL
`of petrol ether were added to the solution before
`centrifugation 3 min at 3500 rpm. The organic
`phase was transferred to a new tube containing 1 g
`anhydrous sodium sulfate. This step was repeated
`and the supernatant evaporated under a nitrogen
`stream. The pellet was recovered in 1.5 mL of
`acetone, dissolved by ultrasound for 30 sec and
`filtered (0.45 µm) prior to HPLC analysis.The HPLC
`system (Surveyor PDA ThermoFisher Scientific)
`was connected to a PDA detector selected to 470
`nm. YMC column 30 (25 cm x 4.6 mm id; particles
`5 microns; pores 100 Å) was preceded by a guard
`
`RIMFROST EXHIBIT 1099 page 0003
`
`Rousset (Université Pierre et Marie Curie-Paris
`6, UMR S872, Les Cordeliers, Paris). They were
`cultured in DMEM Glutamax medium, 4.5 g/L
`glucose, 1 % antibiotics, 1 % nonessential amino
`acids, 20 % inactivated foetal calf serum, in an
`incubator at 37 °C and 10 % CO2. To mimic the
`enterocyte transport, the cells were seeded at a
`density of 0.25 × 106 cells in inserts containing a
`semi permeable PET membrane (23.1 mm in di-
`ameter; 1 µm porosity) placed in 6-well plates
`(Becton Dickinson). The use of inserts allows dif-
`ferential access to the two poles of the cell, the
`apical compartment representing the intestinal
`lumen and basolateral compartment represent-
`ing the internal circulation. In brief, cells were cul-
`tured with complete medium (20% serum) for 7
`days after seeding until it formed a compact cell
`monolayer, as validated by transepithelial electri-
`cal resistance (TEER, Millipore). Cells were kept in
`a serum-free medium on the apical side and com-
`plete medium on the basolateral side for a further
`two weeks to induce differentiation and structural
`configuration similar to physiological conditions.
`The cells were used after 21 days of culture, that is
`to say when they are contiguous and polarized. On
`D20, the culture media were replaced with identi-
`cal media without phenol red. On D21, the apical
`medium was replaced with 500 µl of test medium
`described in section 2.1, while the basolateral me-
`dium was replaced with 1.5 mL complete medium.
`The treatment was stopped after 3 or 6 hours of
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`64
`
`RESULTS AND DISCUSSION
`
`Absorption data
`
`column with the same characteristics. This system
`was kept at a constant temperature of 30 °C during
`elution (constant flow 1 ml/min). After 0.5 min of
`isocratic condition using 4% solvent A (ddH2O),
`15% of solvent B (MTBE) and 81% solvent C
`(Methanol), carotenoids were eluted over 50 min
`with a linear gradient of 15 to 90% of solvent B,
`solvent A remaining constant throughout the
`elution time. The column was re-equilibrated for
`5 min between each analysis. Quantification was
`performed using a standard curve of all-trans
`astaxanthin
`(CaroteNature, Swizerland). The
`integration and analysis of astaxanthin peaks were
`performed using the ChromQuest software.
`All raw data were the product of at least two
`independent experiments. Statistical tests were
`performed using SPSS v17.0 for Windows (SPSS
`Inc). Because of the relatively small number and
`heterogeneity of replicates (n=5 to 13, depending
`on the treatment), all statistical analyses were by
`default non parametric. Kruskal-Wallis tests were
`performed to compare the global sets of data and,
`when significant, post-hoc analyses were done
`using Dunn’s pair to pair comparison.
`The absorption of astaxanthin present in
`the four sources was time-dependent with a me-
`dian value for the unabsorbed fraction of 41.2%,
`53.8%, 48.5%, 35.6% after 3 hours of treatment
`and 15.6%, 44.2%, 2.6%, 6.7% after 6 hours of
`treatmentfor the yeast, algae, bacterium and syn-
`thetic sources, respectively (Fig. 1). Although a
`
`3H
`
`*
`
`MIMOUN-BENARROCH et al.
`lower rate of absorption is observed at 3 hours be-
`tween the esterified and free forms, the difference
`is only significant between algae and synthetic
`astaxanthin (p<0.05). At 6 hours, however, the dif-
`ference is statistically highly significant against all
`three non-esterified sources (p<0.001 for all three
`forms). As a control molecule, beta-carotene had an
`unabsorbed fraction of 31.9% at 3 hours and was
`fully absorbed after 6 hours. This increased rate of
`absorption is not significant at 3 hours compared
`to astaxanthin but is significant at 6 hours against
`yeast (p<0.01) and algae (p<0.001) astaxanthin
`but not the two other sources.Therefore even if
`the transport mechanism is via passive transfer,
`the rate of transfer is different.
`Caco2/TC7 cells were culture for 21 days on
`transwells prior to treatment. Astaxanthin (5µM)
`were added on the apical medium (DMEM serum
`free) and the apical, basal (DMED 10% serum)
`and cell fractions were collected after 3 or 6 hours.
`Carotenoids were quantified and expressed as
`percentage of unabsorbed molecule (apical). All
`datapoints are presented on the scatterplot for
`yeast, algae, bacterium and synthetic sources,
`in order. Dunn’s pair-to-pair comparisons were
`performed for statistical significance. * p<0.05; ***
`p<0.0005Over a third of astaxanthin remained unab-
`sorbed (44.2%) after 6 hours for the algae
`source (Fig. 1). As the percentage of unabsorbed
`astaxanthin is not greatly improved between 3
`and 6 hours of treatment (non-significant), the
`chromatograms confirm that only esterified
`astaxanthin remains in the apical medium while
`
`6H
`
`***
`
`***
`
`***
`
`H.plu
`P.car
`Treatment
`
`H.plu
`P.car
`Treatment
`
`Synth
`
`Fig. 1. Unabsorbed fractions of astaxanthin isoforms after 3 and 6 hours incubation
`
`RIMFROST EXHIBIT 1099 page 0004
`
`P.rho
`
`75
`
`50
`
`25
`
`0
`
`Unabsorbed fraction
`
`(%)
`
`Synth
`
`P.rho
`
`75
`
`50
`
`25
`
`0
`
`Unabsorbed fraction
`
`(%)
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`The Bioavailability of Astaxanthin Is Dependent on Both the Source and the Isomeric Variants of the Molecule
`only free form is present in the cells (data not
`O’Sullivan et al. (2004) showed that astaxan-
`shown). This suggests a role for hydrolysis of
`thin was better accumulated into Caco-2 cells over
`astaxanthin esters leading to delayed absorption
`a 24-hours period compared to beta-carotene,
`of the molecule.
`which is in contradiction with our observations.
`Transfer to the basolateral side was very poor
`However cells were cultured as monolayer rather-
`for astaxanthin (up to 2%, whatever the source)
`than on a transwell system and the delivery matrix
`while it was fast and complete for beta-carotene
`was Tween40/80 rather than biliary micelles. In
`(Tab.2). This difference
`is highly significant
`addition, cells were treated with a mixture of ca-
`(p<0.001) against all sources of astaxanthin at
`rotenoids rather than individual compound, which
`both 3 and 6 hours of treatment. From the different
`may impact the absorption rate of individual mol-
`sources of astaxanthin, there is no molecule
`ecules. A similar result with higher cellular uptake
`detected in the basolateral compartment with
`and secretion to the basolateral side was observed
`algae and the transfer through the basolateral
`on a transwell culture set-up when comparing
`membrane seems to be time-dependent for yeast
`astaxanthin to beta-carotene presented on the
`and synthetic astaxanthin, although very slow, but
`apical side in Tween40 with chylomicron-stimu-
`not for the bacterial astaxanthin. The question
`lating molecules for 16 hours (O’Sullivan, Ryan &
`remains as to why the astaxanthin accumulates
`O’Brien, 2007). Sy et al. (2012) did not find any
`in the cells rather than being exported to the
`carotenoid into the basolateral medium after a 3
`basolateral medium in a similar manner to beta-
`hours incubation in the apical side with natural
`carotene. This may suggest the implication of an
`or synthetic micelles. This is consistent with our
`active transport system.
`observations for astaxanthin (less than 2% in the
`There are very few investigations on the
`basolateral fraction) but not beta-carotene. All ex-
`bioavailability of astaxanthin reported. The main
`perimental conditions were similar between the
`reason may be that records of astaxanthin used as a
`two set-ups. Our data also showed a higher trans-
`bioactive molecule in human nutrition are no more
`fer rate to the cellular fraction at 3 hours (around
`than a couple of decades old. As a consequence,
`50%) compared to the authors (10%) but all agree
`studies of this carotenoid are relatively sparse.
`on a similar uptake rate for the two carotenoids at
`In addition, investigations on the bioavailability
`3 hours. The higher uptake rate should not be due
`represent only a very small part of the available
`to adsorbed carotenoids as the washing step was
`publications, which are mostly on physiological
`performed in our experiments as well.
`benefits. It may be as well due to the highly
`The rate of absorption measured was increa-
`competitive market leading to preclinical research
`sed for the free form compared to esterified astax-
`remaining confidential. However it is possible
`anthin from algae. This is consistent with the data
`to compare our results with those published on
`reported by Lyons et al. (2002), where cellular in-
`other carotenoids.
`corporation from synthetic astaxanthin was nearly
`twice higher than algae astaxanthin. Esterification
`Tab.2. Absorption of astaxanthin and beta-carotene by Caco2/TC7 cells
`3h
`6h
`Source
`of which in BL
`% Absorbed (BL+C)
`of which in BL
`% Absorbed (BL+C)
`87.83 (7.54)
`b
`2.20 (1.69)
`P.rho
`59.22 (6.75)
`1.01 (0.50)
`P.car
`51.94 (13.83)
`1.19 (2.08)
`94.08 (6.58)
`c
`0.75 (1.58)
`H.plu
`45.26 (5.29)
`0.00 (0.00)
`60.20 (9.50)
`a
`0.00 (0.00)
`Synth
`64.38 (7.78)
`1.00 (0.95)
`90.78 (4.70)
`bc
`2.54 (1.70)
`60.32 (15.67)
`55.49 (13.91)
`100.00 (0.00)
`d
`95.87 (6.52)
`B-car
`Caco2/TC7 cells were culture for 21 days on transwells prior to treatment. Astaxanthin or beta-carotene (5µM) were added on the apical medium
`(DMEM serum free) and the apical, basal (DMED 10% serum) and cell fractions were collected after 3 or 6 hours. Carotenoids were quantified
`and expressed as percentage of absorbed molecule (cell + basolateral). Data are expressed as mean and standard deviation and different letters
`in columns indicate a significant difference between the treatments (Kruskall-Wallis followed by Dunn’s pair-to-pair comparisons).
`
`b
`ab
`a
`b
`b
`
`b
`b
`a
`b
`c
`
`65
`
`b
`a
`a
`b
`c
`
`RIMFROST EXHIBIT 1099 page 0005
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`66
`
`is a common feature of lipid-soluble molecules.
`Carboxyl ester lipase, also known as cholesterol
`esterase is produced in the pancreatic juice for an
`activity at the brush border (Ikeda et al., 2003) and
`is the main actor of hydrolysis of carotenoid esters
`in the intestinal lumen before absorption of the
`free form by the enterocytes. However, in a similar
`experimental setup, fucoxanthin was hydrolysed
`when exposed to Caco-2 cells producing the free
`form fucoxanthinol similar to that produced by
`hydrolysis with pancreatic lipase (Sugawaraet al.,
`2002). This activity of endogenous lipase or car-
`boxylesterase activity at the apical membrane or
`secreted into the apical medium close to the brush
`border is consistent with previous reports (Ann-
`aert et al., 1997; Spalingeret al., 1998)”container-
`title”:”Biochimica Et Biophysica Acta”,”page”:”119-
`127”,”volume”:”1393”,”issue”:”1”,”source”:”Pub
`Med”,”abstract”:”Dietary triglycerides, the major
`precursors of long chain fatty acids (FA. Similarly,
`absorption of zeaxanthin, another xanthophyll is
`facilitated by prior hydrolysis of its esters (Chit-
`chumroonchokchai& Failla, 2006). Therefore the
`partial absorption of astaxanthin and detection of
`the free form only in the cellular fraction observed
`herein may be due to the endogenous hydrolytic
`activity of the cells. Further investigations need
`to be performed as well to assess the esterase ac-
`tivity in cells exposed to free and ester forms of
`astaxanthin. In addition, it may be interesting to
`do a pre-digestion with pancreatic juice. The up-
`take and role of esterified astaxanthin on mucosal
`absorption patterns remain unclear at this time.
`Sugawara et al, 2009, observed some esterifica-
`tion of astaxanthin by Caco-2 cells after 24 hours
`treatment. It remained very partial with up to 2%
`of the astaxanthin present in the basolateral com-
`partment being esterified compared to up to 10%
`for peridininol esters. Although we did not detect
`any astaxanthin esters in the basolateral medium,
`our treatment time was much shorter at 3 and 6
`hours and the process may be time-dependent(Sy
`et al., 2012).
`Animal studies (data not shown) have been
`performed in order to better understand the
`impact of the source and the food matrix in
`absorption parameters. A study on hamsters
`showed that powder matrix greatly reduced the
`plasma recovery of astaxanthin from the yeast
`Phaffia rhodozyma and delays the peak of plasma
`concentration by up to 4 hours. This is in agreement
`
`MIMOUN-BENARROCH et al.
`with the clinical study from Mercke-Odeberg et
`al (2003) showing an increased bioavailability
`of astaxanthin from the algae Haematococcus
`pluvialis in lipid compared to powder matrix.
`Another study in rats compared the two main
`sources (algae and yeast) prepared in the same
`lipid carrier (safflower oil). Data showed that
`this specific carrier was not suitable to dissolve
`the free form (yeast) into a solution but rather
`into a fine suspension of astaxanthin crystals.
`Hence, the bioavailability was much lower for the
`yeast compared to algae due to the exchange area
`that was much lower (Ajinomoto Foods Europe,
`personal data).
`These studies were thus contradictory with
`the in vitro data regarding the rate of absorption
`between yeast and algae astaxanthin. However
`two reasons can be suggested for this observation.
`The first is in the formulation for the in vivo stud-
`ies, leading to lower solubility of the free form thus
`lower available concentration of the molecules.
`The second may be related to hydrolysis of astax-
`anthin esters on the intestinal area, possibly due
`to stimulation of the enzyme through the presence
`of the ester, which was not observed in vitro.
`The expression of several
`lipid carriers
`was measured by RT-qPCR after 3 and 6 hours
`of treatment (data not shown). The maximum
`expression was observed at 6 hours, when the
`absorption was mostly complete. The activation
`therefore is not necessarily directly related to an
`active transport of the molecules. Incubation with
`carotenoid-free micelles, a wide range of genes
`were modulated, indicating that most carriers were
`modulated primarily by bile acids, in agreement
`with the physiological mechanisms of absorption
`of dietary lipids. When comparing astaxanthin
`treatments against micelles, very few modulations
`were significant. Astaxanthin from the yeast
`Phaffia rhodozyma stimulated the expression
`of a range of genes at 6 hours, including Fabp2
`(p<0.05), Scarb1 (p<0.05) and Scp2 (p<0.01).
`Astaxanthin from algae Haematococcus pluvialis
`stimulated expression of Scarb1 (p<0.05) and Scp2
`(p<0.05) at 6 hours. Synthetic astaxanthin, finally,
`stimulated the expression of Scp2 (p<0.05) at 6
`hours, while bacterial astaxanthin did not seem
`to significantly affect any gene expression. All this
`may suggest a structure-dependent activation
`
`Mechanisms of action
`
`RIMFROST EXHIBIT 1099 page 0006
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`67
`
`CONCLUSION
`
`The Bioavailability of Astaxanthin Is Dependent on Both the Source and the Isomeric Variants of the Molecule
`of the lipid carriers involved in the uptake or
`and formation of lipid carriers such as micelles for
`at least stimulated by astaxanthin isomers. In
`general lipid uptake as well as intracellular cho-
`the case of Scp2, a possibility may be the role of
`lesterol transport (Shen, Howles, &Tso, 2001).The
`the geometrical isomer cis-astaxanthin which is
`wider effect of the yeast astaxanthin may suggest a
`virtually absent from the bacterial extracts and
`further role in stimulating circulating HDL-choles-
`present in up to 30% in the other forms (Tab. 1).
`terol (SR-B1) or the intestinal recognition protein
`During et al. (2005) in a cell study using the
`and intracellular utilization of long chain fatty ac-
`Caco-2 cell line showed that the bioavailability of
`ids and their esters (FABP2).
`carotenoids was partially inhibited by the use of
`an antagonist of cholesterol transport. Apolar ca-
`rotenoids seemed more sensitive to this inhibition
`We have observed through this series of ex-
`compared to polar carotenoids such as zeaxanthin
`periments variation in the bioavailability and gene
`or lutein. An inhibitor of the SR-B1 transporter
`expression of lipid carriers after treatment for 3
`also had a partial effect on carotenoids uptake,
`to 6 hours of Caco-2/TC7 cells with astaxanthin in
`suggesting a role in active transport of beta-car-
`mixed micelles. Our main conclusion relates to the
`otene. The results of Reboul et al.(2005), suggest-
`delay in uptake due to the esterification of astax-
`ing that lutein is at least partially carried by an
`anthin, the algae source being partially absorbed
`active mechanism involving the SR-B1 receptor.
`when the free form in the yeast or other extracts is
`Further work reviewed in 2013 point to the role of
`nearly completely absorbed after 6 hours of treat-
`SR-B1 carrier protein for a range of carotenes and
`ment. We found very few differences between geo-
`xanthophylls(Reboul, 2013). The corresponding
`metric and optical isomers of astaxanthin, but fur-
`gene, Scarb1 was modulated by both beta-caro-
`ther structural study would be needed to confirm
`tene and some sources of astaxanthin in our study.
`those observations. . Further investigations are
`The inhibition of the expression of AbcA1 is how-
`required as the esterified form from algae is cur-
`ever inconsistent with the putative role of the car-
`rently the only non-dietary source of astaxanthin
`rier protein in the excretion of carotenoids on the
`present for human nutrition and while their physi-
`basolateral membrane of the enterocyte (Reboul,
`ological effects seem to be more and more dissect-
`2013), nor is it consistent with the high transfer
`ed, the first contact with the host at the intestinal
`rate of beta-carotene to the basolateral medium as
`mucosal barrier remains a black box.
`observed herein. Our tests have not to date includ-
`Our second conclusion is based on the ex-
`ed the use of inhibitors of the lipid carriers high-
`pression of genes related to lipid or cholesterol
`lighted in our results or suggested in publications.
`transport, some of which are also known to be in-
`Additional tests will be needed to refine these ob-
`volved in the transport of carotenoids, but the tim-
`servations and to better understand variations be-
`ing seems off between the observed uptake of the
`tween the different sources of astaxanthin.
`molecule and genetic expression. The implications
`The RIVAGE study (Lairon et al., 2009; Vincent
`are perhaps more complex and the RIVAGE study
`et al., 2002)showed a relationship between a poly-
`suggests possible physiological consequences of
`morphism of the gene Scarb1 (SR-B1 transporter)
`modulating these carriers.
`and cardiovascular markers like blood sugar, insu-
`lin and cholesterol levels, while Fabp2 polymor-
`Acknowledgements. This work is part of the
`phism had lower physiological repercussions but
`APHAR project and was financially supported
`mainly on triglycerides. Variations and stimulation
`by the European Regional Development Fund,
`of lipid metabolism genes could have a functional
`1.2.32597 and the Regional Council of Picardie
`impact beyond the active transport of carotenoid.
`(France), 1012011196-11197. The postdoctoral
`As the genetic expression was at the highest once
`fellowship for Dr Mimoun-Benarroch (2011-2012)
`absorption was completed, the transporters may
`was financially supported by the Regional Council
`be involved in facilitated uptake of other lipids
`of Picardie (France), 1012011198. The authors
`rather than astaxanthin itself. With this in mind,
`also would like to thank the competitiveness
`most forms of astaxanthin seem to particularly
`cluster IAR for their support.
`target the intestinal sterol carrier protein (SCP2)
`and modulate bile acid excretion into the lumen
`
`RIMFROST EXHIBIT 1099 page 0007
`
` Bulletin UASVM Food Science and Technology 73(1) / 2016
`
`

`

`68
`
`REFERENCES
`
`1. Ambati R R, Phang S-M, Ravi, S, Aswathanarayana R
`G (2014). Astaxanthin: Sources, Extraction, Stability,
`BiologicalActivities and Its Commercial Applications—A
`Review. Marine Drugs, 12(1), 128–152.
`2. Annaert, P., Kinget, R., Naesens, L., de Clercq, E.,
`&Augustijns, P.
`(1997). Transport, uptake, and
`metabolism of the bis(pivaloyloxymethyl)-ester prodrug
`of 9-(2-phosphonylmethoxyethyl)adenine in an in vitro
`cell culture system of the intestinal mucosa (Caco-2).
`Pharmaceutical Research, 14(4), 492–496.
`3. Armand, M., Borel, P., Pasquier, B., Dubois, C., Senft, M., Andre,
`M., … Lairon, D. (1996). Physicochemicalcharacteristics
`of emulsionsduring fat digestion in humanstomach and
`duodenum. The American Journal of Physiology, 271(1 Pt
`1), G172-183.
`4. Chitchumroonchokchai, C., & Failla, M. L. (2006).
`Hydrolysis of zeaxanthin esters by carboxyl ester lipase
`during digestion facilitatesmicellarization and uptake
`of the xanthophyll by Caco-2 human intestinal cells. The
`Journal of Nutrition, 136(3), 588–594.
`5. Coral-Hinostroza, G. N., Ytrestøyl, T., Ruyter, B., &Bjerkeng,
`B. (2004). Plasma appearance of unesterified astaxanthin
`geometrical E/Z and optical R/S isomers in men given
`single doses of a mixture of optical 3 and 3’R/S isomers of
`astaxanthin fattyacyldiesters. Comparative Biochemistry
`and Physiology. Toxicology&Pharmacology: CBP, 139(1–3),
`99–110.
`6. During, A., Dawson, H. D., & Harrison, E. H. (2005).
`Carotenoid
`transport
`isdecreased and expression
`of the lipidtransporters SR-BI, NPC1L1, and ABCA1
`isdownregulated in Caco-2 cellstreatedwithezetimibe.
`The Journal of Nutrition, 135(10), 2305–2312.
`7. Furr, H. C., & Clark, R. M. (1997). Intestinal absorption
`and tissue distribution of carotenoids. The Journal of
`NutritionalBiochemistry, 8(7), 364–377.
`8. Ikeda, I., Mitsui, K., Matsuoka, R., Hamada, T., Imabayashi, S.,
`Uchino, A., Imaizumi, K. (2003). Cholesterolesterasebound
`to int