Food Research International 38 (2005) 885–891

www.elsevier.com/locate/foodres

Antioxidant activity of chlorophylls and their derivatives

Ursula M. Lanfer-Marquez *, Rosa M.C. Barros, Patricia Sinnecker
Department of Food and Experimental Nutrition, Faculty of Pharmaceutical Sciences, University of São Paulo, Av. Prof. Lineu Prestes,
580, 05508-900, SP, Brazil

Received 1 June 2004; accepted 22 February 2005

Abstract

The antioxidant activity of six natural isolated chlorophyll derivatives and Cu-chlorophyllin was investigated by measuring their
protective action against lipid oxidation. For this, the b-carotene bleaching method and the stable radical 2,2-diphenyl-1-pic-
ryldrazyl (DPPH) scavenging assay were employed. The results obtained by the b-carotene bleaching method showed that all
chlorophyll derivatives presented a dose-dependent response. Pheophorbide b and pheophytin b were the strongest natural antiox-
idant compounds, whose activities were comparable to BHT. The high antioxidant activity found for pheophorbide b, in compar-
ison to pheophorbide a, demonstrated the importance of the aldehyde group for functionality. On the other hand, by the DPPH
assay, all natural pigments showed low antioxidant activity when compared to Trolox. Cu-chlorophyllin, tested by both methods,
presented a higher antioxidant activity than that of natural chlorophylls, showing the importance of the nature of the chelated metal
in the porphyrin ring. The mechanism of antioxidant activity displayed by the natural chlorophyll derivatives does not seem to be
based on the ability to donate hydrogen but maybe, on the protection of linoleic acid against oxidation and/or preventing decom-
position of hydroperoxides.

2005 Elsevier Ltd. All rights reserved.

Keywords: Antioxidant activity; Chlorophyll derivatives; b-carotene bleaching method; DPPH

1. Introduction dant pigments in nature. The scarcity of information

Ingestion of dietary phytochemicals from fruits and
vegetables has proven to play an important role in pre-
venting diseases due to their potent antioxidant activ-
ity. Screening of raw materials for identifying
potential antioxidant activity has resulted in a large
body of published works over the past three decades.
However, there is a gap in knowledge about the health
benefits, if they may either be ascribed to bioactive
phytochemicals like vitamins C and E, carotenoids
and polyphenolics or also attributed to chlorophylls.
Nevertheless, in most research, chlorophylls had not
been included in the trials despite being the most abun-
*
Corresponding author. Tel.: +55 11 3091 3684; fax: +55 11 3815
4410.
E-mail address: lanferum@usp.br (U.M. Lanfer-Marquez).
could be attributed to the difficulty to obtain chloro-
phylls and their derivatives in a purified form and also
to their instability, which triggers a degradation pro-
cess. A research group in Japan (Endo, Usuki, & Kan-
eda, 1985a, 1985b; Usuki, Endo, & Kaneda, 1984b;
Usuki, Suzuki, Endo, & Kaneda, 1984) first suggested
a pro-oxidant activity of chlorophylls under light,
which could be understood as a transfer of the energy
of singlet-excited chlorophyll to oxygen that would
form reactive oxygen species. However, the same
authors also reported that chlorophylls and pheophy-
tins provide protection by preventing autoxidation of
vegetable edible oils stored in the dark and suggested
a hydrogen donating mechanism breaking the radical
chain reactions. In addition, they stated that the intact
chemical structure of porphyrin seems to be essential
for antioxidant activity.

0963-9969/$ - see front matter
2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodres.2005.02.012
886 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
Wanasundara and Shahidi (1998) reported that the
presence of chlorophyll in tea extracts was responsible
for a pro-oxidant effect on the oxidation of marine oils.
Hoshina, Tomita, and Shioi (1998) found that chloro-
phylls were better antioxidants than their metal free
derivatives and confirmed the importance of the porphy-
rin ring on the inhibition of lipid autoxidation. Sakata
et al. (1990) reported that a related compound to pheo-
phorbide a, isolated from clam, appeared to be the most
important antioxidant in the organism. Ferruzzi, Böhm,
Courtney, and Schwartz (2002) provided information
concerning in vitro antioxidant and antimutagenic activ-
ities of water and lipid soluble chlorophyll derivatives as
well as of metal chelated derivatives by free radical scav-
enge and bacterial reverse mutagenesis assays, respec-
tively. The antioxidant activities of the synthetic
metallo-chlorophyll derivatives, especially Cu-chelated
compounds were found to be much higher than those
of natural chlorophylls and of Mg-free derivatives,
which showed an almost insignificant antioxidant capac-
ity. The highest activities were found for Cu-substituted
chlorins. Previously, Sato et al. (1986) identified Cu-
chlorin e4 as the major antioxidant found in commercial
grade sodium Cu-chlorophyllin, by measuring its anti-
oxidant activity in rat liver homogenates.
Questions have been raised about the natural chloro-
phyll metabolites impact on health, due to their high
content in some vegetables, widespread occurrence in
nature, and their degradation during food processing
and human digestion. As very few data are available
on both the antioxidant activity and dose-dependent re-
sponse, the aim of this work was to evaluate not only the
protective action of six natural purified chlorophyll
derivatives (chlorophyll a and b, pheophytin a and b,
pheophorbide a and b) but also the synthetic Cu-chloro-
phyllin against lipid oxidation employing two methods:
the bleaching of b-carotene in a water/linoleic acid
emulsion and the scavenging of the stable radical 2,2-di-
phenyl-1-picrylhydrazyl (DPPH).
2. Materials and methods
2.1. Preparation of pigment standards
All solvents for extraction and chromatography
(acetone, petroleum ether, methanol and water) were
ACS or HPLC grade (Carlo Erba, Milano, Italy).
Commercial antioxidants such as BHT and 6-hydro-
xy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Tro-
lox) were obtained from Sigma–Aldrich (Steinheim,
Germany). Commercial sodium copper chlorophyllin
was obtained from CHR Hansen (São Paulo,
Brazil).
Chlorophylls (a and b) were extracted from 5 g dehy-
drated spinach leaves with cold acetone 80% in a pro-
portion of 1:6 (w/v) and filtered through a sintered
plate funnel under vacuum. The pellet was further ex-
tracted until discoloration of the residue. A separatory
funnel containing the combined supernatants was added
of petroleum ether (about 60 mL) and cold deionized
water (about 20 mL). After vigorous shaking and stand-
ing, the aqueous layer was discarded. The washing pro-
cedure was repeated three to four times to remove
acetone. The ether phase containing the pigments was
dehydrated using sodium sulphate anhydride. This solu-
tion was filtered and concentrated in a rotoevaporator
(Heidolph Elektro GmbH & Co., WB 2000, Kelheim,
Germany). The residue was suspended in 4–5 mL ace-
tone and filtered on a 0.45 lm membrane filters (PTFE
– Millipore) for HPLC analysis (Mangos & Berger,
1997; Schwartz, Woo, & Von-Elbe, 1981).
Chlorophylls (a and b) present in the spinach extract
were separated and collected in a preparative HPLC sys-
tem (Fraction Collector Module – FRC-10A, Shimadzu,
Kyoto, Japan), using a C18 – PREP ODS (20 mm
ID · 25 cm, Shimadzu, Tokyo, Japan) column. The sep-
aration was achieved by applying an isocratic run with
acetone/methanol (80:20 v/v) and a flow rate of
10 mL/min. Fractions of chlorophylls a and b were col-
lected, concentrated under vacuum until dryness, sus-
pended in a known volume of acetone and stored at

20C until analysis.
Pheophytins (a and b) were synthesized from the
collected fractions of chlorophylls (a and b) by drop-
wise addition of 1.0 M HCl (Breemen, Canjura, &
Schwartz, 1991; Schwartz et al., 1981). Conversion
was completed in approximately 2 h in the dark pre-
senting a change of color from green to olive
brown.
The dephytilated derivatives pheophorbides (a and b)
were prepared from the respective chlorophyllides by
acidification with HCl in a similar way as described
for the obtaining of pheophytins from chlorophylls.
Chlorophyllides (a and b) were obtained by the action
of the endogenous chlorophyllase on chlorophylls (a
and b) according to Hörtensteiner, Vicentini, and Matile
(1995) by incubating 5 g of dehydrated spinach with
100 mL acetone/0.2 M Tris–HCl buffer (pH 8.0, 1:1
v:v) at 40 C for 2 h in the dark. Pheophorbides were
separated and collected in the same preparative HPLC
system, as previously mentioned, but with an isocratic
run with methanol/1.0 M ammonium acetate buffer
(pH 7.0)/acetone (80:10:10 v:v:v). All handling proce-
dures were carried out under subdued light and with
cold solvents to avoid degradation.
All pigments separated by preparative HPLC were
concentrated under vacuum. The phytilated chloro-
phylls were resuspended in a small volume of pure
acetone (HPLC grade) and the dephytilated chloro-
phylls in 80% acetone, being both stored at
20 C
until analysis.
 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
2.2. Identification and quantification of isolated pigments
All isolated pigments were identified by analytical
HPLC, according to Mangos and Berger (1997), by
comparison of their retention times and spectral charac-
teristics with those published in the literature (Johnson-
Flanagan & Thiagarajah, 1990) and eventually
confirmed by mass spectrometry.
A triple pump (LC-10ADVP) computer-controlled
(Class-VP 5.032) system (Shimadzu, Tokyo, Japan
CLASS-M10A) and a Nucleosil ODS 100 column
(5 lm, 250 · 4 mm i.d., guard-column 11 · 4 mm i.d.,
Macherey-Nagel, Düren, Germany) with a flow rate of
1 mL/min was used for HPLC analysis. The UV/Vis
diode-array spectrophotometric detector (SPD-
M10AVP) was set simultaneously at 432 nm for chloro-
phylls, 409 nm for pheophytins and pheophorbides
according to their maximal absorbance (Schwartz
et al., 1981).
The solvent system consisted of: (A) methanol/1.0 M
ammonium acetate buffer (pH 7.0) (80:20 v:v), (B) meth-
anol:acetone (80:20 v:v); and (C) acetone. The gradient
started with 100% A to 100% B linearly for 20 min, fol-
lowed by an isocratic hold for 2.5 min, then continued
linearly for 7.5 min to 100% C and was finished by con-
ditioning the column for 4 min with the starting eluent A
(Mangos & Berger, 1997). An aliquot of 10–50 lL was
injected and each sample had three genuine replicates.
Quantifications were made by the calibration curves
of the respectively standards. The calibration curves of
chlorophylls (a and b), pheophytins (a and b) and pheo-
phorbides (a and b) were linear over the range of 2.5–
200 ppm (correlation coefficient = 0.99).
2.3. Determination of antioxidant activity by the b-
carotene bleaching method
For the determination of the antioxidant activity,
the ability to delay the bleaching of b-carotene in a
water/linoleic acid emulsion described by Marco
(1968) modified by Miller (1971) was measured. The
b-carotene undergoes a rapid discoloration in the ab-
sence of an antioxidant and the free linoleic acid rad-
ical attacks the b-carotene molecule, which loses the
double bonds and, consequently, its characteristic
orange color.
Solutions of each natural purified pigment in acetone
(chlorophyll a, chlorophyll b, pheophytin a, pheophytin
b, pheophorbide a, pheophorbide b) and of Cu-chloro-
phyllin in methanol were prepared at 56.7, 113.4,
226.8, 453.7, 680.6 and 907.4 lMol/L. These concentra-
tions correspond to 12.5, 25, 50, 100, 150 and 200 ppm
of BHT equivalents. For comparison, an assay was car-
ried out with butylated hydroxytoluene (BHT). All
determinations were performed in triplicate and results
averaged.
887
For emulsion preparation, 10 mg of b-carotene
(Roche) were dissolved in 10 mL of chloroform. One
milliliter of this solution was then pipetted into an erlen-
meyer containing 20 mg of linoleic acid and 100 mg of
Triton X-100. After the removal of chloroform under
N2, 100 mL of distilled water previously oxygenated un-
der O2 stream for 30 min was dropwise added to the res-
idue with vigorous stirring to form a stable emulsion. A
3 mL aqueous emulsion aliquot was added to 0.2 mL of
each pigment solution in a spectrophotometric cuvette
(light path 10 mm) and the absorbance was immediately
measured at 470 nm, against a blank consisting of the
emulsion without b-carotene. The cuvettes were placed
in water bath at 50 C in the dark and the absorbances
were measured every 15 min until the absorbance had
reached zero (about 2 h).
The antioxidant activity was expressed as percentage
inhibition of oxidation in relation to the control, with-
out any pigment and the decrease of optical density of
the control (D.O.initial
D.O.final) was considered as
100% of oxidation. The decrease of absorbance caused
by the pigments was related to the control and this value
was subtracted from 100, establishing the percentage
inhibition of oxidation.
2.4. Determination of antioxidant activity by the
scavenging of the stable radical DPPH
This assay is based on the measurement of the ability
of antioxidants to scavenge the stable radical 2,2-diphe-
nyl-1-picrylhydrazil (DPPH). The free radical DPPH is
reduced to the corresponding hydrazine when it reacts
with hydrogen donors. For being an easy and accurate
method, it has been recommended to measure the anti-
oxidant activity of fruit and vegetable juices or extracts
(Brand-Williams, Cuvelier, & Berset, 1995; Sánchez-
Moreno, 2002).
A 3 mL aliquot of methanolic DPPH solution
(6 · 10
5 Mol/L) was added to 75 lL of each natural
pigment solution in acetone at a fixed 1 mMol/L con-
centration. The reduction of the amount of DPPH
was followed by monitoring the decrease in absorbance
at 515 nm at 0, 5 and 30 min. Additional tests were made
with Cu-chlorophyllin, Trolox and BHT at 1 mMol/L in
methanol.
The exact initial DPPH concentration (CDPPH) in the
reaction medium was calculated from the calibration
curve over the range from 0.61 to 6.16 · 10
5 Mol/L
by following the equation:
Abs515 nm ¼ 10594  C DPPH
5.2  10
3 ;
as determined by linear regression.
The antioxidant activity was expressed as the inhibi-
tion percentage (IP) of the DPPH radical calculated as
suggested by Moure et al. (2001):
888 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
IP ¼ ðAbst¼0 min
Abst¼30 min Þ=ðAbst¼0 min Þ  100.
Despite the fact that many authors use different reac-
tion times to characterize the antioxidant activity, in this
work a reaction time of 30 min was established, accord-
ing to the original method of Blois (1958). All analyses
were performed in triplicate and results were averaged.
3. Results and discussion
The first step of our study was addressed to obtain
pure pheophytins and pheophorbides, a and b, from
chlorophyll a and b extracted from fresh spinach leaves,
due to the lack of commercially available chlorophyll
derivatives. Extracts of pure chlorophyll (a and b); phe-
ophytin (a and b) and pheophorbide (a and b) were col-
lected by preparative HPLC and their identities were
confirmed by comparison of their spectral characteris-
tics and also by mass spectrometry (Sinnecker, Braga,
Macchione, & Lanfer-Marquez, 2005). Structural char-
acteristics of the natural pigments are shown in Fig. 1
and their spectra, in Fig. 2, being similar to those re-
ported by Johnson-Flanagan and Thiagarajah (1990).
Fig. 3 displays the antioxidant activity of the compo-
nents measured by the b-carotene bleaching method and
results are expressed as percentage inhibition of oxida-
tion in relation to a control without the pigments. The
antioxidant activities were measured over a broad range
of concentrations in comparison to BHT which was cho-
sen due to its wide use to maintain the stability of edible
oils and its high antioxidant activity shown in oil-in-
water emulsions (Becker, Nissen, & Skibsted, 2004; Kol-
eva, Beek, Linssen, Groot, & Evstatieva, 2002).
BHT presented the highest antioxidant activity
among all compounds screened and reached around
90%, which was constant between 50 and 200 ppm.
Pheophorbide b showed the most effective protection
against linoleic acid oxidation. At concentrations from
50 ppm BHT equivalents up, only 20% of b-carotene
was decolorized, while 10% was decolorized by BHT.
Fig. 1. Structure of chlorophyll derivatives assayed for antioxidant activity.
The lower the decolorizing is, the higher the antioxidant
capacity will be. Pheophytin b also showed a relatively
high antioxidant activity, decolorizing about 30% from
50 ppm BHT equivalents up and its profile was similar
to that of pheophorbide b. Despite showing antioxidant
activities as high as BHT, both substances have different
solubilities. While pheophytins are lipophilic, the lack of
phytol in pheophorbides turns the molecule hydrophilic
and therefore, different antioxidant activities are ex-
pected. According to the ‘‘polar paradox’’ which as-
sumes that lipid oxidation occurs at the water/oil
interface, where lipophilic antioxidants are located, the
more hydrophilic pheophorbide would have shown a
lower antioxidant activity than that presented by pheo-
phytin. In the present study, the different solubility
properties responsible for their distribution in the two
phases emulsion seemed not to have affected the effi-
ciency of these two compounds.
Pheophorbide a presented a weak activity, starting
slowly at 12.5 ppm and reaching its highest activity at
150 ppm which was lower than that of pheophorbide b.
Chlorophyll a, the most prominent pigment in green
vegetables, showed the weakest antioxidant activity
among all greenish pigments over the whole range of
concentration studied. A steady increase in activity
was observed by raising its concentration up to
200 ppm BHT equivalents, reaching just 44% of antiox-
idant activity, against 90% observed for BHT. The low
antioxidant activity observed for chlorophyll a may be
explained by its high chemical instability. According to
Usuki et al. (1984) a reaction medium which contains
unsaturated fatty acids promotes the decomposition of
chlorophylls reducing its antioxidant capacity. Unlike
chlorophyll a, the metal-free counterpart pheophytin a,
presented a relatively high protection against oxidation
reaching about 70% between 50 and 100 ppm BHT
equivalents and then a significant decrease occurred at
higher concentrations. This trend of dose–response
dependence has also been observed by other authors
(Moure et al., 2001; Pryzbylski, Lee, & Eskin, 1998).
 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
Fig. 2. Spectra and maximum absorbances (nm) of chlorophyll derivatives.
100
80
Inhibition of oxidation (%)
60
40
20
0
0 25
Concentration (ppm of BHT equivalents)
Fig. 3. Antioxidant activities of chlorophyll derivatives measured by the b-carotene bleaching method in a water/linoleic acid emulsion.
In most cases, the antioxidant activity increases with the
antioxidant concentration up to a certain level, behaving
as prooxidant at higher concentrations.
In contrast to chlorophyll a, the sodium copper chlo-
rophyllin showed a high antioxidant activity, similar to
that of pheophoride b.
The b-carotene bleaching method used is not suitable
to measure antioxidant activity of chlorophyll b. This
pigment absorbs strongly at 462 nm when in acetone
causing a background interference with b-carotene,
which absorbs at 447 and 474 nm. The other pigments
tested did not show a significant interference in the
absorbance readings at 462 nm.
Derivatives b presented higher antioxidant capacity
than derivatives a as a general trend. These results sup-
port the hypothesis that the presence of the aldehyde
group (–CHO) in place of the methyl group (Fig. 1) pro-
vides a better antioxidant activity, although the identifi-
cation of the mechanism involved needs to be
investigated.
In order to compare the results obtained by the b-car-
otene bleaching method with a typical chain breaking
based method, the scavenging of free radical DPPH as-
889
BHT
Cu-chlorophyllin
Pheophorbide b
Pheophytin b
Pheophorbide a
Chlorophyll a
Pheophytin a
50 75 100 125 150 175 200
say was used at a fixed 1 mMol/L concentration for all
pigments, which corresponded to 200 ppm of BHT
equivalents. While Trolox and BHT scavenged about
80% and 43% of DPPH, respectively, all the natural pig-
ments showed low activities, always presenting less than
12% of DPPH radical scavenging (data not shown). In
contrast, Cu-chlorophyllin presented an intermediary
activity quenching around 39% of DPPH, comparable
to the activity of BHT. These results confirm previous
report from Ferruzzi et al. (2002) who determined in vi-
tro antioxidant and antimutagenic activity of chloro-
phyll derivatives by radical scavenging and bacterial
reverse mutagenesis assays. According to the quoted
authors, synthetic metallo-complexes with zinc and cop-
per or commercial grade sodium copper chlorophyllins
(SCC) exhibited antioxidant activity significantly higher
than natural chlorophyll derivatives, showing that the
nature of the chelated metal might alter the ability to do-
nate electrons from the conjugated porphyrin system.
Different results for antioxidant activity of chloro-
phyll derivatives have been obtained by other authors.
Endo et al. (1985a) studied the effects of chlorophylls
and pheophytins in the autoxidation of oils in the dark
890 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
and tested different concentrations of the four pigments,
from 2 to 200 ppm. The strongest antioxidant activity
was found for chlorophyll a, followed by BHT, chloro-
phyll b, pheophytin a and pheophytin b and the four
pigments presented a linear dose-dependent response.
The differences between these results and those in the
present work may be attributed to the different tests
used to evaluate the antioxidant activity. According to
Becker et al. (2004), some factors responsible for obtain-
ing conflicting results in measurements of antioxidant
activity for the same compound are the physical struc-
ture of the test system, the nature of the substrate for
oxidation and the analytical method employed. While
the development of peroxides in a lipophylic system
was used by the cited authors, the bleaching of b-caro-
tene in an aqueous emulsion was used in the present
study. From the foregoing results, it seems that not a
single method is able to offer a comprehensive prediction
of antioxidant efficacy and, therefore, more than one
method should be performed (Aruoma, 2003).
Considering the significant daily dietary intake of
chlorophyll, these results do stimulate further investiga-
tions about natural chlorophylls as antioxidants.
4. Conclusions
Among the natural pigments assayed, pheophorbide
b and pheophytin b showed the strongest antioxidant
activities, which were comparable to BHT, when
tested by the b-carotene bleaching method. Neverthe-
less, despite their potential antioxidant capacity, these
pigments are minor constituents in green vegetables
and do not contribute significantly to the overall pro-
tection against lipid oxidation. Chlorophyll a exhibited
considerable antioxidant activity only at high concen-
trations of about 1 mMol/L, but due to its high level
found in plants, it may play a role in protecting
against lipid oxidation. All natural chlorophyll deriva-
tives had low antioxidant activity when compared to
Trolox according to results obtained by the DPPH as-
say. Hence, the mechanism of action involved might
not be related to the capacity of hydrogen donation
but, to the protection of linoleic acid against oxidation
and/or prevention of the decomposition of hydroper-
oxides. In both methods employed, the Cu-chlorophyl-
lin presented higher antioxidant activity than that of
all natural chlorophylls, showing the importance of
the nature of the chelated metal. More probes based
on different tests should be performed in order to elu-
cidate the mechanism of antioxidant activity of these
compounds. Last but not the least, the potential anti-
oxidant activity of some chlorophyll derivatives is
likely to have important implications regarding health
benefits by diet management and it deserves further
studies.
Acknowledgments
We are grateful to FAPESP (Projects 02/01145-8 and
03/08733-5) and to CNPq (Project 472877/2003-8) for
financial support, and to Cinthya F. Higashi for techni-
cal assistance.
References
Aruoma, O. I. (2003). Methodological considerations for characteriz-
ing potential antioxidant actions of bioactive components in plant
foods. Mutation Research, 523–524, 9–20.
Becker, E. M., Nissen, L. R., & Skibsted, L. H. (2004). Antioxidant
evaluation protocols: food quality or health effects. European Food
Research Technology, 219, 561–571.
Blois, M. S. (1958). Antioxidant determinations by the use of stable
free radical. Nature, 181(26), 1199–1200.
Brand-Williams, W., Cuvelier, M. E., & Berset, C. (1995). Use of free
radical method to evaluate antioxidant activity. Lebensmittel-
Wissenschaft und Technologie, 28, 25–30.
Breemen, R. B., Canjura, F. L., & Schwartz, S. J. (1991). Identification
of chlorophyll derivatives by mass spectrometry. Journal of
Agricultural and Food Chemistry, 39, 1452–1456.
Endo, Y., Usuki, R., & Kaneda, T. (1985a). Antioxidant effects of
chlorophyll and pheophytin on the autoxidation of oils in the dark.
I. Comparison of the inhibitory effects. Journal of the American Oil
ChemistsÕ Society, 62(9), 1375–1378.
Endo, Y., Usuki, R., & Kaneda, T. (1985b). Antioxidant effects of
chlorophyll and pheophytin on the autoxidation of oils in the dark.
II. The mechanism of antioxidative action of chlorophyll. Journal
of the American Oil ChemistsÕ Society, 62(9), 1387–1390.
Ferruzzi, M. G., Böhm, V., Courtney, P. D., & Schwartz, S. J.
(2002). Antioxidant and antimutagenic activity of dietary
chlorophyll derivatives determined by radical scavenging and
bacterial reverse mutagenesis assays. Journal of Food Science,
67(7), 2589–2595.
Hörtensteiner, S., Vicentini, F., & Matile, P. (1995). Chlorophyll
breakdown in senescente cotyledons of rape, Brassica napus L.:
enzymatic cleavage of pheophorbide a in vitro. New Phytology,
129, 237–246.
Hoshina, C., Tomita, K., & Shioi, Y. (1998). Antioxidant activity of
chlorophylls: its structure–activity relationship. Photosynthesis:
Mechanisms Effects, 4, 3281–3284.
Johnson-Flanagan, A. M., & Thiagarajah, M. R. (1990). Degreening
in canola (Brassica napus, c.v. Westar) embryos under optimum
conditions. Journal of Plant Physiology, 136, 180–186.
Koleva, I. I., Beek, T. A., Linssen, J. P. H., Groot, A., & Evstatieva, L.
N. (2002). Screening of plant extracts for antioxidant activity: a
comparative study on three testing methods. Phytochemical anal-
ysis, 13(1), 8–17.
Mangos, T. J., & Berger, R. G. (1997). Determination of major
chlorophyll degradation products. Lebensmittel Unters Forsch A.,
204, 345–350.
Marco, G. J. (1968). A rapid method for evaluation of antioxidants.
Journal of the American Oil ChemistsÕ Society, 45, 594–598.
Miller, H. E. (1971). A simplified method for the evaluation of
antioxidants. Journal of the American Oil ChemistsÕ Society, 45, 91.
Moure, A., Cruz, J. M., Franco, D., Dominguez, M., Sineiro, J.,
Dominguez, H., et al. (2001). Natural antioxidants from residual
sources. Food Chemistry, 72, 145–171.
Pryzbylski, R., Lee, Y. C., & Eskin, N. A. M. (1998). Antiox-
idant and radical-scavenging activities of buckwheat seed
components. Journal of the American Oil ChemistsÕ Society,
75, 1595–1601.
 U.M. Lanfer-Marquez et al. / Food Research International 38 (2005) 885–891
Sakata, K., Yamamoto, K., Ishikawa, H., Yagi, A., Etoh, H., & Ina,
K. (1990). Chlorophyllone-A, a new pheophorbide – a related
compound isolated from Ruditapes philippinarum as an antioxidant
compound. Tetrahedron Letters, 31(8), 1165–1168.
Sánchez-Moreno, C. (2002). Review: methods used to evaluate the free
radical scavenging activity in foods and biological systems. Food
Science and Technology International, 8(3), 121–137.
Sato, M., Fujimoto, I., Sakai, T., Aimoto, T., Kimura, R., & Murata,
T. (1986). Effect of sodium copper chlorophyllin on lipid perox-
idation. IX On the antioxidative components in commercial
preparations of sodium copper chlorophyllin. Chemical and Phar-
maceutical Bulletin, 34(6), 2428–2434.
Schwartz, S. J., Woo, S. L., & Von-Elbe, H. J. (1981). High
performance liquid chromatography of chlorophylls and their
derivatives in fresh and processed spinach. Journal of Agricultural
and Food Chemistry, 29, 533–535.
891
Sinnecker, P., Braga, N., Macchione, E. L. A., & Lanfer-
Marquez, U. M., (2005). Mechanism of soybean (Glycine
max L. Merrill) degreening related to maturity stage and
postharvest temperature of drying. Postharvest Biology and
Technology (accepted).
Usuki, R., Endo, Y., & Kaneda, T. (1984b). Prooxidant
activities of chlorophylls and pheophytins on the photooxi-
dation of edible oils. Journal of Biological Chemistry, 48(4),
991–994.
Usuki, R., Suzuki, T., Endo, Y., & Kaneda, T. K. (1984).
Residual amounts of chlorophylls and pheophytins in refined
edible oils. Journal of the American Oil ChemistsÕ Society,
61(4), 785–788.
Wanasundara, U. N., & Shahidi, F. (1998). Antioxidant and pro-
oxidant activity of green tea extracts in marine oils. Food
Chemistry, 63(3), 335–342.