Astaxanthin diferulate as a bifunctional antioxidant

Article abstract of DOI:10.3109/10715762.2014.982112

Astaxanthin when esterified with ferulic acid is better singlet oxygen quencher with k2 = (1.58 ± 0.1) 10¹⁰ L mol^-1s^-1 in ethanol at 25°C compared with astaxanthin with k2 = (1.12 ± 0.01) 10⁹ L mol^-1s

Full text of DOI:10.3109/10715762.2014.982112

Just Accepted by Free Radical Research
Astaxanthin diferulate as a bifunctional
antioxidant
Thiago Bueno Ruiz Papa, Vagner Dantas Pinho, Eduardo Sanches
Pereira do Nascimento, Willy Glen Santos, Antonio Carlos Bender
Burtoloso, Leif Horsfelt Skibsted, and Daniel Rodrigues Cardoso
doi: 10.3109/10715762.2014.982112
Abstract
Astaxanthin when esterified with ferulic acid is better singlet-oxygen
quencher with k2 = (1.58 0.1) 1010 L mol-1s-1 in ethanol at 25°C
compared to astaxanthin with k2 = (1.12 0.01) 109 L mol-1s-1. The
ferulate moiety is in the astaxanthin diester a better radical scaven-
ger than free ferulic acid as seen from the rate constant of scaveng-
ing of 1-hydroxyethyl radicals in ethanol at 25°C with a second-order
rate constant of (1.68 0.1) 108 L mol-1s-1 compared to (1.60 0.03)
107 L mol-1s-1 for the astaxanthin:ferulic acid mixture, 1:2 equiva-
lents. The mutual enhancement of antioxidant activity for the newly
synthetized astaxanthin diferulate becoming a bifunctional antioxi-
dant is rationalized according to a two-dimensional classification plot
for electron donation and electron acceptance capability.
© 2014 Informa UK, Ltd. This provisional PDF corresponds to the article as it appeared upon acceptance. Fully formatted PDF and
full text (HTML) versions will be made available soon.
DISCLAIMER:The ideas and opinions expressed in the journal’s Just Accepted articles do not necessarily reflect those of Informa Healthcare (the Publisher), the Editors or
the journal.The Publisher does not assume any responsibility for any injury and/or damage to persons or property arising from or related to any use of the material contained
in these articles. The reader is advised to check the appropriate medical literature and the product information currently provided by the manufacturer of each drug to be
administered to verify the dosages, the method and duration of administration, and contraindications. It is the responsibility of the treating physician or other health care
professional, relying on his or her independent experience and knowledge of the patient, to determine drug dosages and the best treatment for the patient. Just Accepted
articles have undergone full scientific review but none of the additional editorial preparation, such as copyediting, typesetting, and proofreading, as have articles published
in the traditional manner. There may, therefore, be errors in Just Accepted articles that will be corrected in the final print and final online version of the article. Any use of the
Just Accepted articles is subject to the express understanding that the papers have not yet gone through the full quality control process prior to publication.

Astaxanthin diferulate as a bifunctional antioxidant
Thiago Bueno Ruiz Papa1, Vagner Dantas Pinho1, Eduardo Sanches Pereira do Nascimento1,
Willy Glen Santos1, Antonio Carlos Bender Burtoloso1, Leif Horsfelt Skibsted2, and Daniel
Rodrigues Cardoso1
1Instituto de Química de São Carlos, Universidade de São Paulo, Av. Trabalhador São
Carlense 400, CP 780, CEP 13560-970, São Carlos – SP, Brazil, 2Department of Food
Science, University of Copenhagen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark.
Corresponding author: Daniel Rodrigues Cardoso, Instituto de Química de São Carlos,
Universidade de São Paulo, Av. Trabalhador São Carlense 400, CP 780, CEP 13560-970, São
Carlos – SP, Brazil. Tel/Fax +55 16 33 73 99 76. E-mail: drcardoso@iqsc.usp.br
Abstract
Astaxanthin when esterified with ferulic acid is better singlet-oxygen quencher with k2 =
o
(1.58 ± 0.1) 1010 L mol-1s-1 in ethanol at 25 C compared to astaxanthin with k2 = (1.12 ±
0.01) 109 L mol-1s-1. The ferulate moiety is in the astaxanthin diester a better radical
scavenger than free ferulic acid as seen from the rate constant of scavenging of 1-
hydroxyethyl radicals in ethanol at 25 oC with a second-order rate constant of (1.68 ± 0.1) 108
L mol-1s-1 compared to (1.60 ± 0.03) 107 L mol-1s-1 for the astaxanthin:ferulic acid mixture,
1:2 equivalents. The mutual enhancement of antioxidant activity for the newly synthetized
astaxanthin diferulate becoming a bifunctional antioxidant is rationalized according to a two-
dimensional classification plot for electron donation and electron acceptance capability.
Keywords: carotenoid, plant phenol, antioxidant, antiradical, astaxanthin.
1

Introduction
Carotenoids are important protectors of biological structures against oxidative stress both as
singlet oxygen quenchers and as radical scavengers [1,2]. Singlet oxygen quenching, as is
important during photosynthesis, depends on energy transfer between the electrophilic singlet
oxygen and the polyene backbone of the carotenoid with rates increasing for increasing
length of conjugation followed by non-radiative/radiative decay of the formed triplet-
carotenoid to the ground state, eq.(1) and eq.(2) [3]:
β-car + 1O2 → 3β-car + 3O2
3β-car → β-car + heat
(1)
(2)
For radical scavenging three mechanisms have now been recognized for β-carotene [4,5]:
β-car + R• → β-car•+ + R-
(3)
(4)
(5)
β-car + R• → β-car-R•
β-car + •OH → β-car(-H)• + H2O
electron transfer eq.(3), radical addition eq.(4) and hydrogen atom transfer eq.(5), for which
the radical scavenging efficiency, as in the reaction eq.(3), depends on the electron donor
capacity [6]. Keto-carotenoids like astaxanthin are poor electron donors and have, based on
theoretical calculations been classified as electron acceptors in their protective functions
during oxidative stress, eq.(6) [8]:
β-car + O2•- → β-car•- + O2
(6)
In a two dimensional donor/acceptor classification, compounds like ascorbic acid are found
among the poor radical scavengers as both being a poor electron donor and a poor electron
acceptor, while β-carotene is an example of good electron donor and at the same time good
electron acceptor and accordingly a good antiradical compound. In this classification,
phenolic antioxidants like the plant phenols and the tocopherols, function as good electron
2

donors but at the same time as poor electron acceptors, while the keto-carotenoids including
astaxanthin and cantaxanthin are good electron acceptors but poor electron donors and have
been termed antireductants [6,8].
It has now been recognized that antioxidant interaction may create synergism between
antioxidants belonging to different groups in the donor/acceptor classification [2]. Such
effects are seen at the interface between lipids and water as in membranes, where lipophilic
carotenoids like β-carotene may protect the lipids, through radical scavenging as in the
reaction of eq.(3) followed by regeneration by plant phenol, Ar-OH, eq.(7):
β-car•+ + Ar-OH → β-car + Ar-O• + H+
(7)
Such regeneration reactions have been found to be fast and important for the antioxidant
synergism observed between compounds both classified as good electron acceptors as for β-
carotene and various plant phenols but with opposite classification as electron donors [9].
Plant phenols covalently linked to carotenoid derivatives have been shown to have better
function in membrane model systems with facile intermolecular electron transfer rather than
electron transfer depending on bimolecular processes between the plant phenols and the
carotenoid [10,11,12,13,14,15].
What remains to be studied is, however, the combination of plant phenols as good electron
donors but poor electron acceptors linked covalently to a poor electron donor but good
electron acceptor such as astaxanthin, Scheme 1. This double opposite combination has only
been studied for bimolecular antioxidant regeneration [16]. The potential of such bifunctional
molecules simultaneously functioning as antireductants and as antioxidants accordingly to the
reactions of eq.(3) and eq.(6) respectively, is completely unexplored and deserves attention
both in relation to singlet oxygen quenching and to radical scavenging activity.
Materials and methods
Chemicals
3

Acetonitrile, dichloromethane (DCM), dimethylformamide (DMF), ethanol (EtOH), ethyl
acetate, hexane, and pyridine (Pyr) were of HPLC grade (J. T. Baker, Phillipsnurg, NJ, US).
Ferric chloride (97%) and hydrogen peroxide (30%) were obtained from Merck (Darmstadt,
Germany). Dichloromethane was dried over 3Å molecular sieves (Sigma-Aldrich, Steinheim,
Gwermany) 24 hours prior to the experiments. Ammonium chloride (99%), cesium carbonate
(Cs2CO3), 4-dimethylaminopyridine (DMAP) (99%), ferulic acid (99%), formic acid (98%),
imidazole (Im) (99%), oxalyl Chloride (COCl)2 (99%), α-(4-pyridyl-1-oxide)-N-tert-
butylnitrone (4-POBN) (99%), tert-butyl dimethyl silyl chloride (TBDMSCl) (95%), sodium
sulfate (99%), and tetrabutylammoniuum hexafluorophosphate were of analytical grade and
purchased from Sigma-Aldrich (Steinheim, Germany). Deionized water was obtained using a
Milli-Q system Millipore Co. (Billerica, MA, US).
Preparation of Ferulic acid TBDMS ether [17]
For the phenolic hydroxyl group with TBDMS 1.70 g (11.3 mmol) of TBDMSCl were added
to a 5 mL stirred CH2Cl2 1 g (5.15 mmol) of ferulic acid, 0.77 g (11.3 mmol) of imidazole
and 0.094 g (0.77 mmol) of DMAP at 0⁰ C. The solution was returned to room temperature
and the reaction was stirred until the reaction was finished. After addition of NH4Cl saturated
solution, the reaction mixture was extracted three times with CH2Cl2 and dried over with
Na2SO4. The organic phase was concentrated and the residue purified by flash column
chromatography on silica gel (ethyl acetate/hexane 2:3) to give 0.93 g (65% yields) of the
desired ferulic acid TBDMS ether.
Preparation of Ferulic acid chloride TBDMS ether [18]
In a solution of 0.233 g (0.84 mmol) equivalent of ferulic acid TBDMS ether in 5 mL of dry
CH2Cl2 under Ar atmosphere, 0.16 g (1.3 mmol) of oxalyl chloride were added dropwise at
0⁰ C. Then, 6 mg (0.084 mmol) of dry dimethylformamide (DMF) was added to the solution
as catalyst. After 2 h of reaction at room temperature, the excess of oxalyl chloride was
removed under reduced pressure and the solvent was removed under vacuum to yield the
ferulic acid chloride TBDMS ether.
Synthesis of Astaxanthin diferulate
0.050 g (0.084 mmol) of astaxanthin was dissolved in 5 mL of dry CH2Cl2 and 250 µL of
pyridine was added under argon atmosphere. The result solution was added dropwise to a
closed vial, purged with Ar, containing a solution of ferulic acid chloride TBDMS ether (0.84
mmol) in 5 mL of dry CH2Cl2 at 0 ºC. The reaction was kept at room temperature overnight
4

under stirring. The reaction solution was dissolved in CH2Cl2 and washed with NH4Cl
saturated solution, the organic phase was dried over with Na2SO4 and concentrated. The
residue was dissolved in 10 mL of DMF/H2O (10:1) solution and 0.274 g (0.84 mmol) of
Cs2CO3 [18] was added under stirring. After 2 h of reaction, water was added and the reaction
mixture was extracted three times with ethyl acetate and dried over with Na2SO4, the organic
phase was concentrated; the residue was diluted in CH2Cl2 and applied to TLC plates.
Purified astaxanthin diferulate were obtained after development with ethyl acetate/hexane
(3:7), to give 0.063 g (yield of 80%) of astaxanthin diferulate.
Transient absorption laser flash photolysis
Laser flash photolysis experiments were carried out with an LFP-112 nanosecond laser flash
photolysis spectrometer from Luzchem (Ottawa, Canada) using the third harmonic (355 nm)
of a pulsed Quantel Nd:YAG laser (Les Ulis, France) attenuated to 14 mJ·cm-2 as the
excitation source with a 8 ns resolution. A R928 photomultiplier tube from Hamamatsu
Photonics (Hamamatsu City, Japan) was used to detect the transient absorption (300-900 nm).
Appropriate UV cutoff filters were used to minimize the sample degradation by the
monitoring light. The samples were excited in 1.0 cm x 1.0 cm fluorescence cuvettes from
Hellma Gmbh. (Mulheim, Germany). Each kinetic trace was averaged 16 times and observed
rate constants were determined by fitting the data with MatLab R2008. All measurements
were made with fresh solutions thermostated at 298 ± 0.5 K and purged with N2 (White-
Martins, Sertãozinho-SP, Brazil) for 60 min before the experiment.
Singlet oxygen scavenging activity
Singlet-excited oxygen lifetime decay was recorded with a time-resolved NIR
spectrofluorometer (Edinburgh Analytical Instruments, U.K.) equipped with a Nd:YAG laser
(Continuum Surelite III), λexc = 532 nm (4th harmonic; pulse 30 ns). The emitted light passed
through silicon and an interference filter and a monochromator before detection with a NIR
photomultiplier (Hamamatsu Co. R5509). The singlet oxygen lifetime was determined by
5

applying first-order exponential fitting to the curve of the phosphorescence decay. Methylene
blue was used as singlet-excited oxygen photosensitizer in ethanol solutions at a
concentration of 1.6 × 10–5 mol L–1. The stock solutions were made dissolving the
compounds in an ethanolic solution of methylene blue (1.6 × 10-5 mol L–1), with
concentrations of 1.0 g/L for the ester and astaxanthin and 3.0 g/L for ferulic acid. The singlet
oxygen quenching experiments were performed by adding different aliquots of stock
solutions directly to the cuvette containing methylene blue (1.6 × 10-5 mol L–1) in ethanol. A
Continuum Surelite III Nd:YAG laser was used as the excitation source operating at 532 nm
(5 ns, 10 Hz). The radiation emitted at 1270 nm was detected at right angles by a liquid
nitrogen cooled photomultiplier from Hamamatsu R5509.
1-hydroxyethyl radical scavenging activity
The formation of 1-hydroxyethyl radicals and spin trapping by 4-POBN were assayed by
mixing 80 μL of a H2O2 solution (6.0 × 10−2 mol L−1) with the reaction mixture containing 1
mL of 4-POBN (3.0 × 10−3 mol L−1) in ethanol contain 10% (v/v) of water, 60 μL of
FeCl2·4H2O (2.0 × 10−3 mol L−1), and varying concentrations of the antioxidant in oxygen
free solution. The incubation time was 1 min at 25 °C, and the reaction was finally freeze-
quenched in liquid nitrogen (77 K) until analysis. The competitive kinetics was conducted by
monitoring the content of 1-hydroxyethyl/4-POBN by ESI-MS [20].
Electrochemical Studies
Cyclic voltammetry was carried out in a PAR model 264A potentiostat (Oak Ridge, TN,
United States) connected to a personal computer using proprietary software for data
acquisition. Electrochemical oxidations were carried out in CH2Cl2 containing 0.1 M of
tetrabutylammonium-hexafluorophosphate as the supporting electrolyte and a three-electrode
system with saturated calomel, a glassy carbon or boron doped electrode for higher potential
window (up to 2 V vs. NHE), and a platinum wire used as reference, work, and auxiliary
electrodes, respectively.
6

High-resolution accurate mass spectrometry
Direct infusion high-resolution accurate ESI-MS spectra of reaction products were performed
on an LTQ-Orbitrap ESI-FT-MS model Velos from Thermo Fisher Scientific mass
spectrometry system (Bremen, Germany) operating in the negative or positive ion detection
mode.
Luminescent Measurements
Fluorescence, phosphorescence and total emission were measurements were carried out using
a Hitachi F-7000 fluorescence spectrometer (Hitachi High-Tech, Tokyo, Japan) at 77 K or
298 K using a quartz cold finger (Wilmad-Labglass, Vineland NJ, US) or a homemade
thermostated cell holder, respectively. Samples were excited in 1.0 cm × 1.0 cm fluorescence
cuvettes from Hellma Gmbh (Mulheim, Germany) for measurement at 298 K or using high-
quality quartz EPR tubes of 4 mm i.d. (Wilmad-Labglass) at 77 K. Fluorescence lifetime
measurements were performed with an Optical Building Blocks Corp. Fluorometer
(Birmingham, U.K.), using the fluorescence time-resolved mode. The excitation and emission
wavelengths were λ = 460 and 530 nm, respectively. Fluorescence decay times were fitted
using a mono-exponential decay function and the best fit obtained by optimized Chi-square
residuals and standard deviation parameters. All solutions were previously deaerated by
purging the cuvette with high-purity N2 (White-Martins, Sertãozinho-SP, Brazil).
DFT calculations
In order to calculate the Rd and Ra indexes [7] eq.(8) and eq.(9), Density Functional Theory as
implemented in Gaussian 03 [21] was employed for all calculations using the B3LYP/6-
31G(d) basis set for full geometry optimization.
(8)
(9)
7

The electrodonating power (ω-) and the electroaccepting power (ω+) were calculated by
eq.(10) and eq.(11), respectively [22]:
(10)
(11)
where VIE is the vertical ionization energy and calculated as the difference between the
energy of the cation and the neutral molecule, assuming that both of these have the ground-
state nuclear configuration of the neutral molecule. VEA is the vertical electron affinity and
calculated as the energy difference between the neutral molecule and the anion, assuming for
both forms the ground-state nuclear configuration of the neutral molecule.
Results and Discussion
Esterification of astaxanthin with ferulic acid according to the reactions of Scheme 1 gave a
compound confirmed by high-resolution accurate electrospray ionization mass spectrometry
and 1D and 2D 1H/13C NMR to be the diester of astaxanthin, see Table 1.
The UV-visible absorption spectrum of the new compound, Figure 1, is seen to be the
superimposed spectrum of the spectra of astaxanthin and the two ferulic acid building blocks
indicating that the conjugation of astaxanthin was not fully enlarged to include the two
ferulate esters. The new ester groups may, however, to some degree be in a keto-enol
tautomeric equilibrium as shown in Scheme 1. Clearly this tautomerism, strongly shifted to
the keto form, is hardly affecting the absorption spectrum, which by spectral Gaussian
deconvolution could be resolved into components corresponding to astaxanthin and ferulic
acid, see insert in Figure 1.
The emission spectrum of astaxanthin was in contrast strongly affected by esterification by
ferulic acid, Figure 2. The fluorescence quantum yield increased by a factor of more than 10
at 77 K following excitation at 280 nm and a factor of 50 at 298 K, see Figure 2A and 2C for
8

fluorescent spectra and Table 2 for numerical values. The phosphorescence was likewise
strongly enhanced and the quantum yield at 77 K increased from 0.03 to 0.23 for excitation at
250 nm with phosphorescence lifetime of 1.6 s for the ester compared to 0.7 s for astaxanthin,
see insert in Figure 2B and 2D. The total phosphorescence of the new compound also shown
in Figure 2C and 2D is remarkable intense for a carotenoid, which normally have low
intersystem crossing yields to populate the triplet state [9]. The triplet state is suggested to
have an increased Φisc for the diferulated as compared to astaxanthin or that the non-
radioactive deactivation is less efficient for the larger molecule.
The decay of the carotenoid triplet state was also followed using transient absorption
spectroscopy, for the chloroform as electron-withdrawing solvent in order to monitor the
radical cation formation from the triplet for both astaxanthin and the diferulate ester, eq.(12),
see Figure 3. For chloroform, the triplet decay is marginally faster than for ethanol as solvent
for both astaxanthin and the diester, which both formed the radical cation. The lifetime of the
radical cation was in the ms time regime and was 20 % shorter for the diester than for
astaxanthin. Notably, the transient triplet absorption spectrum (T-T) is red-shifted
approximately 10 nm for the diferulate compared to astaxanthin in agreement with longer
conjugation of the diester. A comparable red-shift is noted for the radical cations (D-D
transition).
hυ
carotenoid + CHCl3 → carotenoid•+ + CHCl3•- (12)
The reduced energy difference between the lowest triplet state and the ground state for
astaxanthin diferulate as compared to astaxanthin is expected to give a better match to the
singlet oxygen to ground state triplet oxygen transition in effect becoming a better singlet
1
o
oxygen quencher. The second-order rate constants for quenching of O2 in ethanol at 25 C
obtained by time-resolved near-infrared phosphorescence spectroscopy following methylene
blue formation of 1O2 were determined as shown in Figure 4 for astaxanthin, ferulic acid, and
the astaxanthin diferulate and the rate constants present in Table 3. Ferulic acid does hardly
9

quench 1O2, but enhance, when linked as ester, the quenching by astaxanthin by a factor of 10
as seen from the rate constant and also clearly illustrated in Figure 4. The more efficient 1O2
quenching of astaxanthin diferulate is accordingly seen to confirm the importance of lowering
1
the carotenoid triplet state energy for facile energy transfer from O2, making the ester 13-
fold more reactive than free astaxanthin and greatly surpass the reactivity of lycopene (kq =
3.1 1010 L mol-1s-1), the most efficient natural singlet oxygen quencher [3].
Radical scavenging of astaxanthin and astaxanthin diferulate was compared for 1-
k1
hydroxyethyl radical as a moderately reactive carbon-centered radical using a competitive
kinetics approach based on 4-POBN as a spin-trap and probed by electrospray mass
spectrometry [17]:
CH3(•CH)OH + antioxidant → CH3(CH)OH-antioxidant• (13)
k2
CH3(•CH)OH + 4-POBN → CH3(CH)OH-4-POBN•
(14)
The CH3(CH)OH-4-POBN• stable radical adduct was quantified and based on the known
value of k2, the value of was determined for the 3 evaluated antioxidants according:
(15)
where F is the fraction of CH3(•CH)OH trapped by 4-POBN. From the experimental values
shown in Figure 5, the radical scavenging rate constant presented in Table 2 were found.
Again it is seen that the radical scavenging activity of astaxanthin diferulate is enhanced by a
factor of 10 compared with the 1:2 (astaxanthin:ferulic acid) antioxidant mixture.
The reaction of the astaxanthin diferulate with the 1-hydroxyethyl radical was further
studied through product analysis using high resolution accurate ESI-MS. The main reaction
products had the mass of 1037.55 Da, 772.43 Da, and 266.12 Da and were tentatively
identified as shown in Scheme 2. From the reactions in Scheme 2, it is seen that each
astaxanthin diferulate is capable of scavenging four 1-hydroxyethyl radicals and that the
10

radical scavenging is a radical addition centered at the ferulate moieties of the ester rather
than at the polyene backbone.
The standard reduction potential for the two one-electron process related to astaxanthin,
eq.(11) and eq.(12) was in CH2Cl2 as solvent found to have the value of
E1/2 = +1.18V and +0.95V vs. NHE for asta2+ and asta•+ using cyclic voltammetry,
respectively. For the astaxanthin diferulate the values were E1/2 = +1.20V and +0.98V vs.
NHE, respectively, indicating that astaxanthin diferulate is more reducing than astaxanthin.
Ferulic acid in the ester had the value of E1/2 = +1.58V vs. NHE compared to 1.39 V vs. NHE
for the free ferulic acid. Astaxanthin accordingly makes ferulic acid less reducing and ferulic
acid makes astaxanthin more reducing in the ester as compared to the precursors. The ferulate
moiety is the more reducing part of the ester in agreement with addition reaction of 1-
hydroxyethyl radical as shown in Scheme 2.
asta2+ + e- → asta•+
(16)
asta•+ + e- → asta (17)
As seen from the estimative of the standard reduction potentials in the ester as compared to
free ferulate, the electron donating capacity is reduced. Since ferulate is dominating the
radical scavenging activity, of the astaxanthin diferulate, the ester is moving toward the group
of good electron acceptors at the expense of electron donor capacity. The experimental
verification of this change in electronic properties depends on further studies, but therefore
theoretical calculations using the method described by Martinez [7] quantify the change in
eletronic properties in the radical scavenging activity by esterification of ferulic acid with
astaxanthin, see Figure 6.
Conclusions
Astaxanthin is becoming a better singlet oxygen scavenger through esterification with ferulic
acid, at the same time, ferulic acid becomes a better radical scavenger through the
11

esterification. These simultaneous improvements of antioxidant activity for both the plant
phenol and the carotenoid may be explained by the two dimensional electron donor/electron
acceptor capacity diagram seen in Figure 6. Astaxanthin moiety is clearly modified more than
ferulic acid through the esterification with respect to antioxidant properties. For organized
media like cellular membranes, the esterification of carotenoids by phenolic acids may yield
optimal protection through the two antioxidant function unified in the same molecule and
enhance each other also with the perspective of new drug development in relation to
oxidative stress in eyes and skin health.
Acknowledgement
This research is part of the bilateral Brazilian/Danish Food Science Research Program
“BEAM - Bread and Meat for the Future” supported by FAPESP (Grant No. 2011/51555-7
and 2009/54040) to DRC and by the Danish Research Council for Strategic Research (Grant
No. 11-116064) to LHS. DRC thanks CNPq for the research fellowship. Prof. Mauricio da
Silva Baptista (Chemistry Institute, University of São Paulo) is knowledge for grant the
access to the time resolved NIR spectrofluorimeter.
12

References
[1] Böhm F, Edge R, Truscott G. Interactions of dietary carotenoids with activated (singlet)
oxygen and free radicals: Potential effects for human health Molecular Nutrition & Food
Research 2012; 56: 205-216.
[2] Skibsted LH. Carotenoids in antioxidant networks. Colorants or radical scavengers
Journal of Agricultural and Food Chemistry 2012; 60: 2409-2417.
[3] Di Mascio P, Kaiser S, Sies H. Lycopene as the most efficient biological carotenoid
singlet oxygen quencher Archives of Biochemistry and Biophysics 1989; 274: 532-538.
[4] Martinez A, Vargas R, Galano A. What is Important to Prevent Oxidative Stress? A
Theoretical Study on Electron-Transfer Reactions between Carotenoids and Free Radicals.
Journal of Physical Chemistry B 2009; 113: 12113-12120.
[5] Chen, C-H, Han R-M, Liang R, Fu L-M, Wang P, Ai X-C, et al. Direct observation of the
ß-carotene reaction with hydroxyl radical. Journal of Physical Chemistry B 2011; 115: 2082-
2089.
[6] Martinez A. Donator-Acceptor Map and Work Function for Linear Polyene-Conjugated
Molecules. A Density Functional Approximation Study. Journal of Physical Chemistry B
2009; 113: 3212-3217.
[7] Martinez A.; Rodriguez-Girones MA, Barbosa A, Costas M. Donator acceptor map for
carotenoids, melatonin and vitamins Journal of Physical Chemistry A 2008; 112: 9037-9042.
[8] Galano A, Vargas R, Martinez A. Carotenoids can act as antioxidants by oxidizing the
superoxide radical anion. Physical Chemistry Chemical Physics 2010; 12: 193-200.
[9] Liang R, Chen C-H, Han R-M, Zhang J-P, Skibsted LH. Thermodynamic versus Kinetic
Control of Antioxidant Synergism between beta-Carotene and (Iso)flavonoids and Their
Glycosides in Liposomes. Journal of Agricultural and Food Chemistry 2010; 58: 9221-9227.
13

[10] Zeeshan M, Sliwka H-R, Partali V, Martinez A. Electron uptake by classical electron
donators: astaxanthin and carotenoid aldehydes. Tetrahedron Letters 2012; 53: 4522-4525.
[11] Beutner S, Frixel S, Ernst H, Hoffmann T, Hernandez-Blanco I, Hundsdoerfer C, et al.
Carotenylflavonoids, a novel group of potent, dual-functional antioxidants. Arkivoc 2007;
279-295.
[12] Martin H-D, Kock S, Scherrers R, Lutter K, Wagener T, Hundsdoerfer C, et al. 3,3′-
Dihydroxyisorenieratene,
a
Natural Carotenoid with Superior Antioxidant and
Photoprotective Properties. Angewandte Chemie-International Edition 2009; 48: 400-403.
[13] An C-B, Liang R, Ma X-H, Fu L-M, Zhang, J-P, Wang P, et al. Retinylisoflavonoid as
novel membrane antioxidant. Journal of Physical Chemistry B 2010; 114: 13904-13910.
[14] Hundsdoerfer C, Stahl W, Mueller TJJ, De Spirt S. UVA Photoprotective Properties of
an Artificial Carotenylflavonoid Hybrid Molecule. Chemical Research in Toxicology 2012;
25: 1692-1698.
[15] Hu F, Bu YZ, Liang R, Duan RM, Wang S, Han RM, et al. Quercetin and daidzein ß-
apo-14'-carotenoic acid esters as membrane antioxidants. Free Radical Research 2013; 47:
413-421.
[16] Han R-M, Chen C-H, Tian Y-X, Zhang J-P, Skibsted LH. Fast Regeneration of
Carotenoids from Radical Cations by Isoflavanoid Dianions: Importance of the Carotenoid
Keto Group for Electron Transfer Journal of Physical Chemistry A 2010; 114: 126-132.
[17] Fache F, Suzan N, Piva O. Total synthesis of cimiracemate B and analogs. Tetrahedron
2005; 61: 5261-5266.
[18] Santos S, Graca J. Glycerol ϖ-hydroxyacid-ferulic acid oligomers in cork suberian
structure Holzforschung 2006; 60: 171-177.
14

[19] Jiang ZY, Wang YG. A mild, efficient and selective deprotection of t-
butyldimethylsilyl-protected phenols using cesium carbonate. Tetrahedron Letters 2003; 44:
3859-3861.
[20] de Almeida NEC; Homem-de-Mello P, De Keukeleire D, Cardoso DR. Reactivity of
Beer Bitter Acids Toward the 1-Hydroxyethyl Radical as Probed by Spin-Trapping Electron
Paramagnetic Resonance (EPR) and Electrospray Ionization− Tandem Mass Spectrometry
(ESI− MS/MS). Journal of Agricultural and Food Chemistry 2011; 59: 4183-4191.
[21] Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al.
Gaussian 03; Gaussian, Inc: Wallingford, CT, 2004.
[22] Gazquez JL, Cedillo A, Vela A. Electrodonating and electroaccepting powers. Journal of
Physical. Chemistry A 2007; 111: 1966-1970.
15

Table Legends
Table 1. NMR data collected for astaxanthin diferulate in DMSO-d6 at 25 °C.
1H
13C
( )
Position
, multiplet, (J COSY HSQC
HMBC
in Hz)
1, 1´
2, 2´ eq
2, 2´ ax
36.7
42.1
-
-
C1, 1´
C2, 2´
-
2.04-2.14 m
H3, 3´
C1, 1´, C3, 3´; C4, 4´, C6, 6´, C16, 16´
5.56 dd (12.9;
3, 3´
70.4
H2, 2´
C3, 3´
C1, 1´, C2, 2´, C1´´, 1´´´, C4, 4´
6.3)
4, 4´
5, 5´
193.6
127.0
160.5
123.4
141.8
135.0
134.8
-
-
C4, 4´
C5, 5´
C6, 6´
C7 7´
C8, 8´
C9, 9´
-
-
-
-
-
6, 6´
-
-
7, 7´
6.32 d (16.3)
6.57 d (16.3)
-
H8, 8´
H7, 7´
-
C1, 1´, C5, 5´, C8, 8´, C9, 9´, C10, 10´
C6, 6´, C9, 9´, C10, 10´, C19, 19´
-
8, 8´
9, 9´
10, 10´
6.46 d (11.8)
H11, 11´ C10, 10´
H10,
C8, 8´, C12, 12´, C19, 19´
11, 11´
125.0
6.67-6.83 m
;
10´
C11, 11´
C9, 9´, C10, 10´, C13, 13´
H12, 12´
H11, 11´ C12, 12´
C13, 13´
H15, 15´ C14, 14´
H14, 14´ C15, 15´
C9, 9´, C10, 10´, C13, 13´, C14, 14´
,
12, 12´
139.3
6.51 d (14.8)
C20, 20´
13, 13´
14, 14´
15, 15´
16, 16´
17, 17´’
18, 18´
19, 19´
20, 20´
1´´, 1´´´
2´´, 2´´´
136.5
133.7
130.9
25.8
-
6.39-6.43 m
6.76-6.83 m
1.37 s
-
-
C15, 15´
C13, 13´, C14, 14´
-
-
-
-
-
-
C16, 16´
C17, 17´
C18, 18´
C19, 19´
C20, 20´
C1´´, 1´´´
C1,1´, C2,2´, C6,6´, C17,17´
29.9
1.22 s
C1,1´, C2,2´, C3,3´, C6,6´, C16,16´
C4, 4´, C5, 5´, C6, 6´, C8, 8´, C16,16´
C8, 8´, C9, 9´, C10, 10´
13.8
1.83 s
12.2
2.01 s
12.5
1.97 s
C12, 12´, C13, 13´, C14, 14´
,
165.7
114.0
-
-
6.57 d (15.8)
H3´´, 3´´´ C2´´, 2´´´
C3, 3´, C1´´,1´´´, C4´´, 4´´´, C5´´, 5´´´
1´´, 1´´´, C2´´, 2´´´, C4´´, 4´´´, C5´´, 5´´´
C6´´, 6´´´, C9´´, 9´´´
C
,
3´´, 3´´´
145.6
7.60 d (15.8)
H2´´, 2´´´ C3´´, 3´´´
4´´, 4´´´
5´´, 5´´´
6´´, 6´´´
7´´, 7´´´
8´´, 8´´´
9´´, 9´´´
125.4
111.1
147.9
149.4
115.4
123.2
-
-
-
-
-
C4´´, 4´´´
C5´´, 5´´´
C6´´, 6´´´
C7´´, 7´´´
-
7.36 s
C3´´, 3´´´, C7´´, 7´´´, C9´´, 9´´´
-
-
-
-
6.80 d (8.2)
7.14 d (8.2)
H9´´, 9´´´ C8´´, 8´´´ C5´´, 5´´´, C4´´, 4´´´, C6´´, 6´´´, C7´´, 7´´´
H8´´, 8´´´ C9´´, 9´´´
C3´´, 3´´´, C5´´, 5´´´
C10´´,
10´´´
10´´, 10´´´ 55.6
3.83 s
-
C5´´, 5´´´, C6´´, 6´´´
7´´, 7´´´-
-
9.63 brs
-
-
-
OH
16

Orbitrap FT-MS ESI+: calculated for C60H69O10 949.48907; found 949.49261 error of 4.3
ppm
17

Table 2. Photophysical properties of ferulic acid, astaxanthin and astaxanthin diferulate in
ethanol.
Compounds
Photophysical
astaxanthin
diferulate
properties
ferulic acid
astaxanthin
226, 290, 316,
345, 478
abs (nm)
F (nm)
230, 289, 318
290, 390
240, 294, 329, 345, 478
322, 334, 347
285, 310, 360
M-1 cm-1
95,800 (486 nm); 54,600
(329 nm)
S
22,000 (318 nm)
125,000 (472 nm)
( )
(
exc=250
0.010 (77 K);
0.002 (298 K)
0.015 (77 K);
0.004 (298 K)
0.190 (77 K); 0.180 (298
K)
F
nm)
P (77K)
P (77K)
máx (T-T)
0.001
0.003
0.7 s
0.230
1.6 s
−
−
−
450 nm
5.2 µs
440 nm
4.1 µs
T
máx (radical)
(radical)
−
860 nm
2.4 ms
850 nm
2 ms
−
ES1
ET1
3.46 eV
3.35 eV
4.33 eV
3.87 eV
4.16 eV
3.54 eV
abs= absorption wavelength, F = fluorescence emission wavelength, S = molar extinction
coefficient,
= fluorescence quantum yield,
= excitation wavelength,
=
P
F
exc
phosphorescence quantum yield,
= phosphorescence lifetime,
(T-T) = maximum
máx
P
absorption wavelength for the triplet-triplet transition
= triplet state lifetime,
T
máx
(radical) = maximum absorption wavelength for the dublet-dublet transition, (radical)
= radical lifetime, ES1 = singlet excited state energy, ET1 = triplet excited state energy
18

Table 3. Second-order rate constants for 1-hydroxyethyl radical scavenging and singlet-
o
excited oxygen deactivation by selected antioxidants and mixture at 25 ± 0.1 C in ethanol
solutions.
1-hydroxyethyl radical
k2’ / (108 L mol-1s-1)
1O2
Antioxidant
kq /( 1010 L mol-1s-
1
)
astaxanthin diferulate
astaxanthin
ferulic acid
1.68 ± 0.10
0.18 ± 0.02
0.02 ± 0.01
0.16 ± 0.03
1.58 ± 0.10
0.12 ± 0.01
not observed
-
astaxanthin:ferulic acid (1:2)
19

Figure Legends
Figure 1. Normalized absorbance spectra of astaxanthin, ferulic acid and
astaxanthin diferulate in ethanol at room temperature. Absorbance
normalized for ferulic acid at maximum 320 nm and for astaxanthin and
astaxanthin diferulate at 472 nm. Insert: UV band of astaxanthin diferulate
resolved into components using spectral Gaussian deconvolution.
20

Figure 2. Luminescence spectrum of astaxanthin and astaxanthin diferulate
in ethanol at 77 K and 298 K. A) Total emission of astaxanthin at 298 K and
77 K in ethanol. B) Fluorescence and phosphorescence spectra of
astaxanthin, inset display the time trace for the phosphorescence decay at 77
K in ethanol. C) Total emission of astaxanthin diferulate at 298 K and 77 K
in ethanol. D) Fluorescence and phosphorescence spectra of astaxanthin
diferulate, inset display the time trace for the phosphorescence decay at 77 K
in ethanol. Arrows indicate the corresponding y-axis (fluorescence intensity
at 298K or total emission at 77K) for the plotted spectra.
A
B
1000
800
600
400
200
0
2000
1500
1000
500
0
2000
1500
1000
500
0
astaxanthin
77K
astaxanthin
Fluorescence at 298 K
300
250
200
150
100
τdecay= 0.7 s
298 K
Fluor. + Phosp. (77K)
Phosph. (77K)
50
0
0
1
2
3
4
5
Ti me (s)
260 280 300 320 340 360 380 400 420 440
250
300
350
λ (nm)
400
450
λ (nm)
C
D
2000
astaxanthin diferulate
77K
1600
1400
1200
1000
800
600
400
200
0
astaxanthin diferul ate
1500
1000
8000
Flu orescence (298K)
Phosph. (77K)
Fl + Phosph. (77K)
298 K
τdecay = 1.6 s
1500
1000
500
0
500
0
6000
4000
2000
0
0
2
4
6
8
10
Time (s)
300
350
400
450
500
300
350
400
450
λ (nm)
λ (nm)
21

Figure 3. Transient difference absorbance decay observed at different
wavelengths upon excitation with 8 ns laser pulses (14 mJ cm2) at 355 nm of
Ar saturated CH2Cl2 solutions of 1.0 x 10-6 mol L-1 astaxanthin (upper
panels) and 1.0 x 10-6 mol L-1 astaxanthin diferulate (lower panels). Each
trace is the average of 16 independent decays at 25 oC.
A
B
0.012
0.010
0.008
0.006
0.004
0.002
0.000
-0.002
0.010
0.008
0.006
0.004
0.002
0.000
λ = 450 nm
λ = 860 nm
astaxanthin
λ = 860 nm
τ = 2.4 ms
astaxanthin
τ = 5.2 μs
0
5
10
15
20
0
2
4
6
8
10
time (ms)
time (μs)
C
D
0.010
0.008
0.006
0.004
0.002
0.000
0.010
0.008
0.006
0.004
0.002
0.000
λ = 440 nm
λ = 850 nm
astaxanthindiferulate
astaxanthin diferulate
λ = 850 nm
τ = 2.0 ms
τ = 4.1 μs
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
time (μs)
time (ms)
22

Figure 4. Plot of experimentally determined rate observed rate constant (kobs)
for singlet oxygen as function of the antioxidant concentration. Second-order
rate constant for singlet-excited oxygen deactivation is obtained from the
slope of the linear regression.
23

Figure 5.
Plot of
vs. the antioxidant
concentration as obtained by spin-trapping ESI-MS.
24

Figure 6. Donator-acceptor map for selected antioxidants including the
astaxanthin direfulate and its precursors based on the theoretically calculated
-
+
electrodonating (ϖ ) and eletroaccepting (ϖ ) power in relation to Na and F
respectively [6].
25

Scheme 1. Synthesis of astaxanthin diferulate. Reagents and conditions: (a)
Im, DMAP, TBDMSCl, DCM; (b) (COCl)2, DMF(cat); (c) Pyr, DCM and (d)
Cs2CO3, DMF:H2O (1:10).
26

Scheme 2. Proposed reaction mechanism for 1-hydroxyethyl radical
scavenging by astaxathin diferulate based on high resolution accurate ESI--
MS data.
27