Hydroperoxy-arachidonic acid mediated n-hexanal and (Z)-3- and (E)-2-nonenal formation in Laminaria angustata

Article abstract of DOI:10.1016/S0031-9422(03)00026-8

In higher plants, C6 and C9 aldehydes are formed from C18 fatty acids, such as linoleic or linolenic acid, through formation of 13- and 9-hydroperoxides, followed by their stereospecific cleavage by fatty acid hydroperoxide lyases (HPL). Some marine algae can also form C6 and C9 aldehydes, but their precise biosynthetic pathway has not been elucidated fully. In this study, we show that Laminaria angustata, a brown alga, formed C6 and C9 aldehydes enzymatically. The alga forms C9 aldehydes exclusively from the C20 fatty acid, arachidonic acid, while C6 aldehydes are derived either from C18 or from C20 fatty acid. The intermediates in the biosynthetic pathway were trapped by using a glutathione/glutathione peroxidase system, and subjected to structural analyses. Formation of (S)-12-, and (S)-15-hydroperoxy arachidonic acids [12(S)HPETE and 15(S)HPETE] from arachidonic acid was confirmed by chiral HPLC analyses. These account respectively for C9 aldehyde and C6 aldehyde formation, respectively. The HPL that catalyzes formation of C9 aldehydes from 12(S)HPETE seems highly specific for hydroperoxides of C20 fatty acids.

Full text of DOI:10.1016/S0031-9422(03)00026-8

Phytochemistry 63 (2003) 669–678
www.elsevier.com/locate/phytochem
Hydroperoxy-arachidonic acid mediated n-hexanal and (Z)-3- and
E)-2-nonenal formation in Laminaria angustata
(
a,1
a
a
b
Kangsadan Boonprab , Kenji Matsui , Yoshihiko Akakabe , Norishige Yotsukura ,
a,
Tadahiko Kajiwara *
aDepartment of Biological Chemistry, Faculty of Agriculture, Yamaguchi University, Yamaguchi, 753-8515, Japan
bInstitute of Algological Research, Faculty of Science, Hokkaido University, Hokkaido 051-0003, Japan
Received 29 October 2002; received in revised form 6 January 2003
Abstract
In higher plants, C6 and C9 aldehydes are formed from C18 fatty acids, such as linoleic or linolenic acid, through formation of
3- and 9-hydroperoxides, followed by their stereospecific cleavage by fatty acid hydroperoxide lyases (HPL). Some marine algae
1
can also form C6 and C9 aldehydes, but their precise biosynthetic pathway has not been elucidated fully. In this study, we show that
Laminaria angustata, a brown alga, formed C6 and C9 aldehydes enzymatically. The alga forms C9 aldehydes exclusively from the
C20 fatty acid, arachidonic acid, while C6 aldehydes are derived either from C18 or from C20 fatty acid. The intermediates in the
biosynthetic pathway were trapped by using a glutathione/glutathione peroxidase system, and subjected to structural analyses.
Formation of (S)-12-, and (S)-15-hydroperoxy arachidonic acids [12(S)HPETE and 15(S)HPETE] from arachidonic acid was con-
firmed by chiral HPLC analyses. These account respectively for C9 aldehyde and C6 aldehyde formation, respectively. The HPL
that catalyzes formation of C9 aldehydes from 12(S)HPETE seems highly specific for hydroperoxides of C20 fatty acids.
#
2003 Elsevier Ltd. All rights reserved.
Keywords: Laminaria angustata; Laminariaceae; Brown algae; Lipoxygenase–hydroperoxide lyase pathway; (S)-12-Hydroperoxy arachidonic acid;
(
S)-15-Hydroperoxy arachidonic acid; n-Hexanal; Nonenal
1
. Introduction
the (phyto)oxylipin or octadecanoid pathway. The
pathway starts from fatty acids, and various products
such as jasmonates, hydroxylated fatty acids, keto-fatty
acids, and long chain aldehydes, are formed in addition
to the C6 and C9 aldehydes. In Phaseolus vulgaris (L.)
leaves, C6 aldehydes are rapidly formed during hyper-
sensitive resistance response against phytopathogenic
Pseudomonas syringae (Croft et al., 1993). In Arabi-
dopsis, it has been shown that the aldehydes can induce
a subset of defense-related genes (Bate and Rothstein,
1998). From these results it has been established that the
aldehydes have significant roles in resistance response of
higher plants against abiotic and biotic attacks. The
molecular mechanism of the pathway is now being
clarified. Lipoxygenases form 13- and/or 9-hydroper-
oxides of linoleic or linolenic acid. Fatty acid hydro-
peroxide lyase, a novel type of cytochrome P450
C6 and C9 aldehydes are volatile compounds that
have a fresh green and cucumber-like flavor. In essential
oil prepared from edible brown algae, Laminaria spp.,
C9 aldehydes such as (E)-2-nonenal (1) could be found
in relatively large quantities, followed by C6 aldehydes,
such as n-hexanal (2), (Z)-3- and (E)-2-hexenal (Kaji-
wara et al., 1996). In higher plants, C6 and C9 alde-
hydes are formed from unsaturated fatty acids, such as
linoleic or linolenic acid through sequential reactions by
lipoxygenase (LOX) and fatty acid hydroperoxide lyase
´
(HPL) (Blee, 1998). This enzymatic pathway is a branch
of a highly divergent biosynthetic pathway, called either
*
Corresponding author. Tel./fax: +81-839-33-5849.
E-mail addresses: ffisksb@nontri.ku.ac.th, boonprabk@hotmail.
(CYP74), acts on the hydroperoxides. From 13-hydro-
com (K. Boonprab), kajiwara@agr.yamaguchi-u.ac.jp (T. Kajiwara).
1
peroxides, C6 aldehydes and C12 oxo acids are formed,
whereas from 9-hydroperoxides, C9 aldehydes and C9
oxo acids are formed. This type of HPL is sometimes
called as a heterolytic HPL. Most HPLs characterized so
Present address: Department of Fishery Products, Faculty of
Fisheries, Kasetsart University, 50 Phaholyothin Rd Chatuchak,
Bangkok 10900, Thailand. Tel.: +66-29428644, +66-29428645; fax:
+66-29428363.
0031-9422/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0031-9422(03)00026-8

670
K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
far, show high substrate-specificity to 13-hydroperoxides.
However, another type of HPL that can act on both the
distillation extraction (Kajiwara et al., 1996). The alde-
hydes other than C6 and C9 are also thought to be formed
from fatty acids; however, their biosynthetic pathway
has not yet been completely elucidated. After incubation
1
3- and 9-hydroperoxides at almost equal efficiency has
been found recently (Matsui et al., 2000). In addition, C5
or C8 compounds are also formed from 13-hydroper-
oxides, presumably through homolytic cleavage of the
hydroperoxides. Thus, the HPL accounting for C5 or
C8 formation is termed a homolytic HPL (Matsui,
ꢀ
of the homogenized fronds at 5 C for 80 min, the quan-
tity of almost all compounds including (E)-2-nonenal
(1) and n-hexanal (2) increased, which suggested that
these compounds were enzymatically formed from
endogenous substrates during incubation. In higher
plants it is well established that these C6 and C9 alde-
hydes are formed from fatty acids such as linoleic and
linolenic acids, through sequential catalyses by LOX
1
998). On the other hand, the C6 and C9 aldehyde-
forming pathway in marine algae has not yet been fully
studied. Recently, it has been shown that a diatom
(Thalasssiosira rotula) can form short-chain aldehydes
having C10 structures, such as (E, Z)-2,4-decadienal
and (E, Z, Z)-2,4,7-decatrienal, which inhibit egg clea-
vage of copepods (Ban et al., 1997). They are thus
believed to be involved in chemical defense in the dia-
tom (Pohnert, 2002). C10 aldehydes are proposed to be
formed from arachidonic acid through formation of 11-
hydroperoxyeicosatetraenoic acid formed by LOX fol-
lowed by cleavage by HPL (Pohnert and Boland, 1996;
Pohnert, 2000). Because marine algae contain relatively
high quantities of C20 and C22 fatty acids, and some
can be used as substrates by LOX, it was thought that
these C20 or C22 fatty acids would also serve as sub-
strates of the lyase branch to form C6 and C9 aldehydes
and heterolytic HPL (Blee, 1998). Interestingly, the
´
amounts of 1-octen-3-ol and 1-octen-3-one also increased
after incubation. These C8 compounds are known to be
formed via pyrolysis of hydroperoxides of arachidonic
acid, and, in fungi and mushrooms, they are shown to
be formed from 10-hydroperoxide of linoleic acid by
homolytic HPL (Wurzenberger and Grosch, 1984a, b,
1986). Because heat was avoided during incubation and
extraction, biosynthetic enzymes might be responsible
for the increase of the C8 compounds in the homo-
genate. This suggests that heterolytic and homolytic
HPLs are present in the fronds of L. angustata.
(
Kajiwara, 1997). In this case, it is assumed that hydro-
peroxides of C20 and C22 fatty acids, should be formed
Gerwick, 1994). This accumulating knowledge led us to
2.2. Unsaturated fatty acid precursors for the aldehyde
branch
(
further attempt to elucidate the enzymatic process forming
C6 and C9 aldehydes in marine algae. In this study we
elucidated that C9 aldehydes, (Z)-3-nonenal (3) and (E)-2-
nonenal (1), and C6 aldehyde, n-hexanal (2), could be
formed from arachidonic acid (4) through formation of
regio- and stereo-specific formation of hydroperoxides
of the fatty acids. Detailed structural analyses of the
hydroperoxides were also carried out.
Previously it was suggested that C20 fatty acids, such
as arachidonic acid (4), might be a precursor for the
formation of C6 and C9 aldehydes in marine algae
(Kajiwara, 1997). This is in contrast to the aldehyde-
forming pathway in higher plants that predominantly
uses C18 fatty acids, such as linolenic and linoleic acids
(Hatanaka 1996; Matsui, 1998; Noordermeer et al.,
2001). In order to confirm the substrates for C6 and C9
aldehyde formation in marine algae, both arachidonic
2
. Results and discussion
Table 1
Composition of volatile compounds identified by SPME-GC–MS
from crude homogenate prepared from L. angustatab
2.1. Enzymatic C6 and C9 aldehyde-formation in
homogenized fronds
Content (mg/100 g fr. wt)
Compounds
Without incubation
With incubation
In order to investigate the enzymatic formation of C9
and C6 aldehydes, their amounts were quantified with
SPME-GC–MS after incubating the homogenate of
n-Hexanal (2)
(E)-2-Hexenal
1-Octen-3-one
29.7
131
69.8
216
n.d.a
ꢀ
fronds at 5 C for 80 min. With this method, volatiles
ꢀ
98.5
44.7
37.1
n.d.a
868
(E)-2-Octenal
1
114
120
were collected at 40 C, so that very little pyrolysis
-Octen-3-ol
could occur if heat-labile hydroperoxides were in the
homogenate. In the SPME extract from the homogenate
of L. angustata, (E)-2-nonenal (1) was found as the
main product followed by (E,Z)-2,6-nonadienal, 1-
octen-3-one, and (E)-2-nonen-1-ol (Table 1). C6 alde-
hydes, such as (E)-2-hexenal and n-hexanal (2) could
also be found, as well as the other C7, C8, C12, and C14,
which were aldehydes as previously found in simultaneous
(
(
(
E, E)-2, 4-Heptadienal
E)-2-Nonenal (1)
E, Z)-2, 6-Nonadienal
38.8
1093
242
175
(E)-2-Nonen-1-ol
E, E)-2, 4-Dodecadienal
Tetradecanal
82.7
67.3
52.6
119
103
(
33.5
a
n.d.: not detected.
Homogenized fronds were incubated at 4 C for 80 min.
b
ꢀ

K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
671
Fig. 1. Substrate specificity for n-hexanal (2), (Z)-3-nonenal (3) and (E)-2-nonenal (1) formation in the homogenate prepared from L. angustata.
ꢀ
Reaction of crude enzyme (2.9 mg protein) with unsaturated fatty acid (700 nmol) was performed at 4 C for 30 min. Under the condition employed,
aldehyde-formation proceeded almost linearly until 30–40 min. Each reaction was done in duplicate, with triplicate HPLC analyses. The means of
duplicates are shown. The amounts of aldehydes formed from with (&) and without (&) exogenous fatty acids were determined by reversed phase
HPLC.
(
4) and linoleic acid were added to the crude enzyme
lian cells (Bryant et al., 1982; Hatzelmann et al., 1989;
Weitzel and Wendel, 1993; Schnurr et al., 1996; Suther-
land et al., 2001). Since fatty acid hydroxides cannot be
substrates for HPL, the addition of the system to the
reaction mixture would inhibit formation of C6 and C9
aldehydes. This procedure has been successfully used to
confirm the involvement of a fatty acid hydroperoxide
as an intermediate in an oxylipin pathway (Hamberg et
al., 1998, 1986; Hamberg and Gerwick, 1993; Bro-
dowsky et al., 1992; Hombeck et al., 1999). When GSH
was added to the reaction mixture, a slight increase in
the amount of aldehyde was observed (data not shown).
It seemed that GSH enhanced the aldehyde-forming
reaction probably through protecting the enzymes from
oxidative inactivation. This result also suggested that
the activity of endogenous GPx in the homogenate, if
any, was negligible. Addition of GPx showed essentially
little effect on the activity, which in turn suggested that
there was little endogenous GSH in the reaction mix-
ture. As shown in Fig. 2, addition of GPx to the reac-
tion mixture containing 3 mM GSH decreased the
activity to form both the C6 and C9 aldehydes from
arachidonic acid (4) in a concentration-dependent man-
ner. This result strongly indicates that there must be the
formation of hydroperoxides of arachidonic acid (4) as
intermediates to form C6 and C9 aldehydes. This sug-
gests that LOX–HPL system(s) exist in the marine algae.
solution and the amount of C6 and C9 aldehydes were
quantified. Formation of these aldehydes proceeded
almost linearly until 30–40 min after addition of the
fatty acid. As shown in Fig. 1, formation of C6 and C9
aldehydes was hardly detected without the crude
enzyme solution, which excluded the possibility of
nonenzymatic formation through autoxidation and
subsequent non-enzymatic cleavage of the fatty acids. It
was found that C9 aldehydes (1 and 3) were essentially
formed from arachidonic acid (4), while the formation
of C9 aldehydes (1,3) from linoleic acid was relatively
low. In contrast, an activity to form n-hexanal (2) was
low, although a very slight increase in the amount could
be detected either from linoleic or arachidonic acid (4)
with almost the same effectiveness.
2
.3. Hydroperoxy fatty acids as the intermediates
According to current knowledge of the aldehyde-
forming pathway in higher plants, the results shown
above suggested that C9 aldehydes (1 and 3) might be
formed from the 12-hydroperoxide of arachidonic acid
(5), and at the same time, the 15-hydroperoxide of ara-
chidonic acid (6) might also be formed to support the
C6 aldehyde formation from arachidonic acid (4). It can
be assumed that LOX(s) forms the hydroperoxides from
arachidonic acid, and that they are further cleaved by
heterolytic HPL(s). In order to examine whether
hydroperoxides of arachidonic acid are formed as
intermediates during C6 and C9 aldehyde formation,
glutathione–glutathione peroxidase (GSH–GPx) system
was used. This system is known to reduce lipid hydro-
peroxides to the corresponding hydroxides in mamma-
2.4. Structural analyses of hydroxides of arachidonic
acid
The previous experiment indicated that hydroper-
oxides of arachidonic acid (4) were formed as inter-
mediate of the HPL branch in the oxylipin pathway in

672
K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
Fig. 2. Effect of glutathione–glutathione peroxidase on activities to form C6 and C9 aldehydes from arachidonic acid (4). Reaction of crude enzyme
ꢀ
(2.6 mg protein) with arachidonic acid (4) (700 nmol) was performed at 4 C for 30 min in the presence of a given amount of glutathione peroxidase
(0–2.5 unit/ml) and 3 mM glutathione. Relative amounts (%, the amounts without the peroxidase was assigned as 100%) of n-hexanal (2) (*) and a
sum of (Z)-3-nonenal (3) and (E)-2-nonenal (1) (*) are shown.
the marine algae. It would be plausible that, in the pre-
sence of GSH and GPx, the hydroperoxides formed as
the intermediates must be trapped and converted into
the corresponding hydroxides. Thus, arachidonic acid
trace b) i.e. their retention times coincided with those of
the peaks of compounds II and III found in the reaction
mixtures. These coincidences were further confirmed by
co-injection, in which compound II coeluted with
15HETE and compound III coeluted with 12HETE. In
order to confirm the structure, the hydroxy fatty acids
purified by HPLC were methylated, hydrogenated and
converted into their TMSi ether derivatives, then they
were subjected to GC–MS analyses. The mass spectrum
of compound II (15HETE) showed prominent peaks at
m/z (rel. int.): 173 (100), 343 (37.5), 367 [M-47] (7), 383
[M-31] (3.1), 399 [M-15] (1.5). Whereas compound III
(12HETE) showed prominent peaks at m/z (rel. int.):
215 (100), 301 (90.6), 367 [M-47] (9.4), 383 [M-31] (3.1),
399 [M-15] (1.6). The characteristic ion peaks at 173 and
343 (m/z) and at 215 and 301 (m/z) are diagnostic for
the hydroxyl groups of 15HETE and 12HETE, respec-
tively (Boeynaems et al., 1980). These evidences sup-
ported that the two compounds are formed from the
intermediates, 12-hydroperoxy-5(Z),8(Z),10(E),14(Z)-
eicosatetraenoic acid (5) and 15-hydroperoxy-
5(Z),8(Z),11(Z),13(E)-eicosatetraenoic acid (6). On the
other hand, compound I is still unidentified.
(4) was incubated with the crude enzyme solution in the
presence of GSH and GPx. Fatty acids and their deri-
vatives were then extracted and analyzed with normal
phase HPLC employing a photodiode array detector.
Two major absorbing compounds with an lmax at 235
nm could be detected in the complete system (Fig. 3).
These two compounds were thought to be the reduced
form of the hydroperoxides formed as intermediates
from arachidonic acid (4).
In order to analyze their structures, the hydroxides
trapped by the GSH and GPx system were isolated in
larger amounts, partially purified, and analyzed with
reversed phase HPLC. Under these HPLC conditions,
three unknown peaks, which were tentatively named as
compounds I, II, and III, were resolved (Fig. 4, trace a)
and all had lmax values of 227, 234, and 234 nm,
respectively. When authentic samples of 12-hydroxy-
5
(Z),8(Z),10(E),14(Z)-eicosatetraenoic acid (12HETE)
and 15-hydroxy-5(Z),8(Z),11(Z),13(E)-eicosatetraenoic
acid (15HETE) were subjected to the HPLC analyses,
they eluted at 11.5 and 12.7 min, respectively (Fig. 4,
To determine the stereochemistries of the HETE
methyl esters, each was subjected to a chiral phase

K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
673
Fig. 3. Enzymatic generation of hydroxy-arachidonic acids in L. angustata. Reaction was composed of (A) crude enzyme (1.9 mg protein), arachi-
donic acid (4) (700 nmol), glutathione peroxidase (4 unit/ml) and glutathione (3 mM), (B) as in A but without arachidonic acid (4), or (C) as in A but
ꢀ
without crude enzyme solution. Each reaction was performed at 4 C for 30 min. A chromatogram with an internal standard (15-hydroxy-
eicosadienoic acid) is showed in panel D.
HPLC system. Under the chiral HPLC condition
employed, both the enantiomers of 12 and 15HETE could
be baseline separated (Fig. 5). By using enantiomerically
pure S-isomer of each hydroxide, it has been revealed that
the (R)-isomer of each elutes faster than the correspond-
ing (S)-isomer. When each hydroxide derived from the
products formed by the algae were subjected to the HPLC
system, one peak could be found respectively, and each
retention time coincided with the respective (S)-isomer.
This was again confirmed by co-injection, and these
results indicated that 12HETE and 15HETE formed by
the algal enzyme had their hydroxyl group in (S)-con-
figuration with optical purities over 99% e.e.
3. Concluding remarks
C6 aldehydes [n-hexanal (2), (Z)-3- and (E)-2-hexenal]
are ubiquitously found in higher plants. In general, they
are formed from C18 fatty acids, such as linoleic or

674
K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
Fig. 4. Reversed phase HPLC analysis of hydroxy-arachidonic acids. Reaction was composed of crude enzyme (240 mg protein), arachidonic acid
4) (7 mmol), glutathione peroxidase (4 unit/ml) and glutathione (3 mM). The products isolated from the reaction mixture (trace a), and mixture of
(
authentic 15HETE and 12HETE (trace b) were analysed. Trace c is a chromatogram obtained with co-injection of reaction mixture and authentic
standards.
Fig. 5. Determination of configurations and optical purities by a chiral phase HPLC of methyl-15HETE (A) and methyl-12HETE (B) formed from
arachidonic acid by crude enzyme solution. The peaks corresponding to authentic enantiomers were indicated with arrows.
linolenic acid, through the formation of 13-hydroper-
oxides, followed by their stereospecific cleavage by 13-
HPL. Some plants, such as pear, cucumber, or melon
can form both C9 and C6 aldehydes (Noordermeer,
2001). C9 aldehydes are also formed from C18 fatty
acids through the formation of 9-hydroperoxides from
the latter. It has been reported that some marine algae
might also be able to form C6 and C9 aldehydes (Kaji-
wara et al., 1996). However, their biosynthetic pathway
has not been elucidated fully. In this study, we show
that L. angustata forms the aldehydes enzymatically.
The algae forms C9 aldehydes (1, 3) almost exclusively

K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
675
from C20 fatty acid (4), arachidonic acid, through the
formation of 12(S)HPETE (5), while C6 aldehydes (2)
are derived from either C18 or C20 fatty acids, through
formation of 13(S)-hydroperoxyoctadecadienoic acid
mostly from Wako Chemical Co., Japan. (Z)-3-Nonenal
(3) was synthesized as described elsewhere (Kajiwara et
al., 1975). 15-Hydroxyeicosadienoic acid (15HEDE)
was prepared as described (Martini et al., 1994; Ruster-
´
(
Boonprab et al., 2003) or 15(S)HPETE (6) (Fig. 6). The
ucci et al., 1999). All solvents used in experiments were
purified by distillation. For quantification experiments,
treatments were prepared in duplicate, and HPLC ana-
lyses were repeated three times.
HPL that accounts for the formation of C9 aldehydes
from 12(S)HPETE (5), seems to be the one highly spe-
cific to hydroperoxides of C20 fatty acids, because par-
tially purified HPL, which is responsible for C6
aldehyde formation, shows little activity to hydroper-
oxides of C20 fatty acids (Boonprab et al., 2003).
4.2. Plant materials
L. angustata was harvested at Charatsunai beach,
Hokkaido (northern part of Japan) facing the Pacific
Ocean during January and February, 2001. Fronds
were kept at 4 C immediately after the harvest and the
temperature was maintained during shipping. For solid
phase micro extraction analysis they were kept frozen
4
. Experimental
ꢀ
4
.1. General
ꢀ
Chemicals: arachidonic acid (4) (590% purity), lino-
at À80 C. For biochemical experiments, they were
ꢀ
leic acid (595% purity), GSH and GPx were purchased
from Sigma Chemical Co., USA. 12(R,S)HETE,
kept at À80 C as coarse powder prepared with liquid
N . The whole powder was mixed prior to taking a
2
12(S)HETE, 15(R,S)HETE and 15(S)HETE were from
Cayman Chemical Co., USA. Other chemicals were
sample (20 g) for each experiment in order to avoid
individual differences.
Fig. 6. The proposed pathway for C6 (n-hexanal (2)) and C9 [(Z)-3-nonenal (3) and (E)-2- nonenal (1)] formation in brown alga, L. angustata.

676
K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
4
.3. Solid phase micro extraction (SPME)
The ether solution was subjected to a prep. TLC
(
Merck, silica gel 60 F-254, 10Â20 cm, EtOAc:n-hexane
L. angustata (2 g fr. wt) was cut into small pieces and
homogenized in 50 mM MES–KOH buffer (pH 5.5, 10
ml) with a Polytron mixer (Kinematca, Switzerland).
2:1). The yellow bands corresponding to the authentic
compounds (R 0.6–1.0) were scraped off and extracted
f
with distilled Et O (2Â2 ml). The ether extract was fil-
2
ꢀ
The homogenate was incubated at 5 C for 80 min, and
tered using a hydrophobic filter unit (0.5 mm DISMIC-
3JP, Advantech TOYO, Japan), and evaporated with a
then, solid CaCl2 (660 mg) was added to inactivate
enzymes. As a control, fronds were added to MES–
KOH buffer (10 ml) containing CaCl (660 mg). The
ꢀ
vacuum centrifuge at 32 C; the resultant residues were
then dissolved in CH CN (50 ml) and subjected to
2
3
mixture was then homogenized with a Polytron mixer.
A portion of the mixture (20 ml) was transferred to a 22
ml screw-cap vial and 1.0 mM n-octanal (20 ml) was
added as an internal standard. Sampling of the head
space was performed by inserting a fiber (Carbowax-
divinylbenzene, 65 mm, Supelco, Bellefonte, PA. USA)
quantification by reversed phase HPLC. Quantification
of the aldehydes was carried out by using five points of
a calibration standard curve constructed with various
amounts of authentic aldehydes (n-hexanal (2), (Z)-3-
nonenal (3) or (E)-2-nonenal (1)) based on the area ratio
of the respective peak of aldehyde to that of the internal
standard (n-heptanal). The linear regression of the three
standard curves was 50.9950. Reversed phase HPLC
was performed on Zorbax SB C18 (250Â4.6 mm, 5 mm)
with a flow rate of 1 ml/min (CH CN:H O:THF
66:33:1(v/v/v)), and UV detection at 350 nm (Shimadzu,
LC-9A, Japan). The activity to form aldehydes varied
from fronds to fronds depending on the collection date,
the site, and the developmental stage of the specimens
(Boonprab et al., unpublished data; Kajiwara et al.,
ꢀ
for 30 min at 40 C. The adsorbed compounds were
analyzed on GC–MS equipped with a fused silica capil-
lary column (60 mÂ0.25 mm) coated with DB-WAX
(
0.25 mm), using helium as a carrier gas. The column
temperature was programmed to increase from 40 C (2
3
2
ꢀ
ꢀ
ꢀ
min hold) to 240 C at 10 C/min. The ionization energy
was 70 eV.
4
.4. Crude enzyme preparation
1993; Sekiya et al., 1984); however, the profiles (i.e.,
L. angustata was crushed into a fine powder under
substrate and product specificities of enzyme system)
shown in this paper were practically reproducible.
Representative data are shown.
liquid nitrogen with a mortar and pestle, and subse-
quently with a Maxim homogenizer (Nihonseiki Kaisha
Ltd., Japan). The powder was transferred into a glass
bottle containing three volumes (v/w) of 0.1 M borate
borax buffer pH 9.0, containing 2% Polyclar VT. The
suspension was homogenized using a Polytron mixer to
break down polysaccharides, then filtered through six
4.6. Analyses of hydroxy-arachidonic acids
To the chilled substrate solution, GSH and GPx were
added to concentration of 3 mM and 4 unit/ml, respec-
tively, from which crude enzyme solution (1 ml) was
added. The enzymatic solution was then incubated at
layers of cheese cloth. The filtrate was centrifuged at
ꢀ
3
076 g for 15 min at 4 C to remove debris. The pH of
ꢀ
the supernatant was readjusted to 6.9 on ice.
Protein contents were determined with the modified
method of Lowry (Dulley and Grieve, 1975).
4 C for 30 min, following which 1 M citric acid (150 ml)
was added to adjust the pH to 3–4. Oxygenated fatty
acids were extracted with Et O (2Â2 ml), with combined
2
Et O extracts washed with satd. NaCl (2 ml), followed
2
ꢀ
4
.5. Analyses of aldehydes
by drying over Na SO at À10 C for 3 h. The Et O
2
4
2
solutions were transfered to a new tube and evaporated
until dry under a gentle stream of N2 gas, whereas
15HEDE (50 ml of 2.25 mM dissolved in n-hexane) was
added to the residue. Separation of hydroxyeicosa-
tetraenoic acid(s) [HETE(s)] was carried out by normal
phase HPLC on Zorbax-SIL 250Â4.6 mm with a flow
rate of 1 ml/min [n-hexane:iso-PrOH:HOAc 987:12:1
(v/v)]. Detection was performed with a photodiode
array detector (Shimadzu, LC-10ADvp, Japan).
A buffer (1 ml, 0.1 M MES-KOH, pH 6.9 prepared
with seawater) and 700 nmol of arachidonic acid (4) (20
mM dissolved in EtOH) or linoleic acid (50 mM dis-
solved in EtOH) in a test tube was chilled on ice for 15
min then, crude enzyme solution (1 ml) was added to it.
ꢀ
This mixture was incubated at 4 C with gentle shaking.
A preliminary experiment showed that the reaction was
linear with time until 30–40 min; hence, in this study,
the reaction was terminated by adding 1% 2,4-DNPH
(
in EtOH containing 0.5 M HOAc, 1.4 ml) and 1 mM
4.7. Analyses of hydroxy-arachidonic acid for structure
and stereoisomer identification
n-heptanal (8 ml) as the internal standard after 30 min.
The hydrazone derivatives were extracted with n-hexane
(
removal of n-hexane with a vacuum centrifuge to yield a
5 ml) and washed with satd. NaCl (2 ml). After
The reaction of crude enzyme and arachidonic acid in
the presence of GPx was scaled-up five-fold, with deri-
vatives of fatty acids being extracted using a Sep-pak
yellow powder, it was then re-dissolved in Et O (50 ml).
2

K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
677
ꢀ
C18 silica cartridge (Water corporation Milford, Mas-
sachusetts USA) (Lehmann et al., 1992). The acids were
next eluted by MeOH, and following evaporation of
to increase from 150 to 200 at 5 C/min. Sample injec-
tion was carried out with a split ratio of 1/50. The ioni-
zation energy was 70 eV.
MeOH, the resultant yellow oil was re-dissolved in Et O
2
and washed with satd. NaCl. The Et O layer was next
2
ꢀ
dried (Na SO , À10 C) for 12 h, with the Et O solubles
Acknowledgements
2
4
2
removed, and the residue subjected to prep. TLC essen-
tially as described above. HETE(s) were separated using
a solvent system of n-hexane:Et O:HOAc 1:1:0.001, v/v.
This work was performed under JSPS-NRCT Core
University Program on ‘‘Development of thermo-
tolerant microbial resources and their applications’’
under the cooperation of Japanese and Thai scientists,
in association with Kasetsart University (Thailand) and
Yamaguchi University (Japan). This study was sup-
ported in part by the San-Ei Gen Foundation for Food
Chemical Research (2002).
2
The spot at R 0.13–0.2 was scraped off, and HETE(s)
f
were extracted from the gel with Et O (5Â2 ml). After
2
removal of ether, the residue was re-dissolved in Et O (5
2
ml) and washed with satd. NaCl soln. The ether layer
ꢀ
was next dried (Na SO at À10 C for 12 h), with the
2
4
solvent then removed and the residue reconstituted in
EtOH (1 ml). The concentration of HETE(s) were esti-
mated by reading A at 234 nm by using an extinction
À1
À1
coefficient of 23200 M cm (Graff et al., 1990). The
soln. was next diluted with EtOH to give 50 mM HETE
and stored at À80 ꢀC until needed. Reversed phase
HPLC was performed using a Zorbax SB C18 (250Â4.6
References
Ban, S.H., Burns, C., Castel, J., Chaudron, Y., Christou, E., Escri-
bano, R., Umani, S.F., Gasparini, S., Ruiz, F.G., Hoffmeyer, M.,
Ianora, A., Kang, H.K., Laabir, M., Lacoste, A., Miralto, A., Ning,
X., Poulet, S., Rodriguez, V., Runge, J., Shi, J., Starr, M., Uye, S.I.,
Wang, Y., 1997. The paradox of diatom–copepod interactions. Mar.
Ecol. Prog. Ser. 157, 287–293.
mm, 5 mm) column, eluted with (MeOH:H O:HOAc
2
8
0:20:0.01 (v/v/v)) at a flow rate of 1 ml/min, and UV
detection at 234 nm (Shimadzu, LC-9A, Japan). Frac-
tions containing 12HETE and 15HETE were collected
following reversed phase HPLC and individually con-
centrated using a Sep-pak C18 cartridge. Aliquots were
methylated with ethereal diazomethane for normal
phase HPLC analyses, using a Zorbax-SIL 250Â4.6 mm
column eluted with n-hexane:iso-PrOH 100:3 (v/v) at a
flow rate of 1 ml/min. Chiral phase HPLC was performed
with Chiralcel-OD-H 250Â4.6 mm at a flow rate of 0.5
ml/min (n-hexane: iso-PrOH 99:1 (v/v)) for 12HETE, or at
a flow rate of 0.25 ml/min (n-hexane:iso-PrOH ; 100:3 (v/
v)) for 15HETE.
Bate, N.J., Rothstein, S.J., 1998. C
ygenase pathway induce a subset of defense-related genes. Plant J.
6, 561–569.
e, E., 1998. Phytooxylipins and plant defense reactions. Prog. Lipid
6
-volatiles derived from the lipox-
1
Ble
´
Res. 37, 33–72.
Boonprab, K., Matsui, K.,Yoshida, M., Akakabe, Y., Chirapart, A.,
Kajiwara, T., 2003. C6-aldehyde formation by fatty acid hydroper-
oxide lyase in brown alga Laminaria angustata. Z. Naturforsch. 56c,
2
07–214.
Brodowsky, I.D., Hamberg, M., Oliw, E.H., 1992. A linoleic acid
8R)-dioxygenase and hydroperoxide isomerase of the fungus
(
Gaeumannomyces graminis. Biosynthesis of (8R)-hydroxylinoleic
acid and (7S, 8S)-dihydroxylinoleic acid from (8R)-hydroperoxy-
linoleic acid. J. Biol. Chem. 267, 14738–14745.
Bryant, R.W., Simon, T.C., Bailey, J.M., 1982. Role of glutathione
peroxidase and hexose monophosphate shunt in the platelet lipoxy-
genase pathway. J. Biol. Chem. 257, 14937–14943.
4
.8. MS Analyses of hydroxy-arachidonic acid
For MS analysis the procedure done by Lehmann et
Boeynaems, J.M., Brash, A.R., Oates, J.A., Hubbard, W.C., 1980.
Preparation and assay of monohydroxy-eicosatetraenoic acids.
Anal. Biochem. 104, 259–267.
al. (1992) was followed. After purification of the
HETE(s), they were methylated with ethereal diazo-
methane. The sample was re-dissolved in EtOH (800 ml),
and platinum oxide (6 mg) was added. Hydrogen gas
was bubbled through a sample soln. for 90 min, with
platinum oxide being removed by filtration. The solvent
was removed under stream of nitrogen, to which they
then added BSTFA soln. (200 ml, Sigma-Aldrich). The
¨
Croft, K.P.C., Juttner, F., Slusarenko, A.J., 1993. Volatile products of
the lipoxygenase pathway evolved from Phaseolus vulgaris (L.)
leaves inoculated with Pseudomonas syringae pv phaseolicola. Plant
Physiol. 101, 13–24.
Dulley, J.R., Grieve, P.A., 1975. Simple technique for eliminating
interference by detergents in the Lowry method of protein determi-
nation. Anal. Biochem. 64, 136–141.
ꢀ
Gerwick, W.H., 1994. Structure and biosynthesis of marine algal oxy-
lipins. Biochim. Biophys. Acta 1211, 243–255.
mixture was next incubated for 60 min at 60 C with
constant stirring under argon. After cooling to room
temperature, the reagents were removed under a gentle
stream of nitrogen. The sample was re-dissolved in n-
hexane (40 ml) and subjected to GC–MS analysis. GC–
MS (GCMS-QP5050A, Japan) was equipped with a
fused silica capillary column (60 mÂ0.25 mm) coated
with DB-WAX (0.25 mm film thickness) using helium as
a carrier gas. The column temperature was programmed
Graff, G., Anderson, L.A., Jaques, L.W., 1990. Preparative and pur-
ification of soybean lipoxygenase-derived unsaturated hydroperoxy
and hydroxy fatty acids and determination of molar absorptivities
of hydroxy fatty acids. Anal. Biochem. 188, 38–47.
Hamberg, M., Herman, C.A., Herman, R.P., 1986. Novel biological
transformations of 15-Ls-hydroperoxy-5,8,11,13-eicosatetraenoic
acid. Biochim. Biophys. Acta 877, 447–457.
Hamberg, M., Gerwick, W.H., 1993. Biosynthesis of vicinal dihydroxy
fatty acids in the red alga Gracilariopsis lemaneiformis: identification

678
K. Boonprab et al. / Phytochemistry 63(200 3) 669–678
of a sodium-dependent 12-lipoxygenase and a hydroperoxide iso-
merase. Arch. Biochem. Biophys. 305, 115–122.
Pohnert, G., 2000. Wound-activated chemical defense in unicellular
planktonic algae. Angew. Chem. Int. Edit. 39, 4352–4354.
Pohnert, G., Boland, W., 1996. Biosynthesis of the algal pheromone
hormosirene by the fresh-water diatom Gomphonema parvulum
(Bacillariophyceae). Tetrahedron 52, 10073–10082.
Hamberg, M., Su, C., Oliw, E., 1998. Manganese lipoxygenase. Dis-
covery of a bis-allylic hydroperoxide as product and intermediate in
a lipoxygenase reaction. J. Biol. Chem. 273, 13080–13088.
Hatanaka, A., 1996. The fresh green odor emitted by plants. Food
Rev. Int 12, 303–350.
2
Pohnert, G., 2002. Phospholipase A activity trigger the wound-acti-
vated chemical defense in the diatom Thalassiosira rotula. Plant
Physiol 129, 103–111.
Rusterucci, C., Montillet, J.L., Agnel, J.P., Battesti, C., Alonso, B.,
´
Hatzelmann, A., Schatz, M., Ullrich, V., 1989. Involvement of glu-
tathione peroxidase activity in the stimulation of 5-lipoxygenase
activity by glutathione-depleting agents in human polymorpho-
nuclear leukocytes. Eur. J. Biochem. 180, 527–533.
Knoll, A., Bessoule, J.J., Etienne, P., Suty, L., Blein, J.P., Trianta-
phylides, C., 1999. Involvement of lipoxygenase-dependent produc-
tion of fatty acid hydroperoxides in the development of the
hypersensitive cell death induced by cryptogein on tobacco leaves. J.
Biol. Chem. 274, 36446–36455.
Hombeck, M., Pohnert, G., Boland, W., 1999. Biosynthesis of dic-
tyopterene A: stereoselectivity of a lipoxygenase/hydroperoxide
lyase from Gomphonema parvulum (Bacillariophyceae). Chem.
Commun 243–244.
¨
Schnurr, K., Belkner, J., Ursini, F., Schewe, T., Kuhn, H., 1996. The
Kajiwara, T., 1997. Dynamic studies on bioflavor of seaweed. Koryo
selenoenzyme phospholipid hydroperoxide glutathione peroxidase
controls the activity of the 15-lipoxygenase with complex substrates
and preserves the specificity of the oxygenation products. J. Biol.
Chem. 271, 4653–4658.
196, 61–70 (in Japanese with English summary).
Kajiwara, T., Kodama, K., Hatanaka, A., Matsui, K., 1996. Biogen-
eration of volatile compounds via oxylipins in edible seaweeds. In:
Takeoka, G.R., Teranishi, R., Williams, P.J., Kobayashi, A. (Eds.),
Biotechnology for improved foods and flavors. American Chemical
Society Symposium series 637, Washington, DC, pp. 146–166.
Kajiwara, T., Matsui, K., Hatanaka, A., Tomoi, T., Fujimura, T.,
Kawai, T., 1993. Distribution of an enzyme system producing sea-
weed flavor: conversion of fatty acids to long-chain aldehydes in
seaweeds. J. appl. Phycol 5, 225–230.
Sekiya, J., Kajiwara, T., Hatanaka, A., 1984. Seasonal change in
activities of enzymes responsible for the formation of C6-aldehydes
and C6-alcohols in tea leaves, and the effects of environmental
temperatures on the enzyme activities. Plant Cell Physiol. 25, 269–
280.
Sutherland, M., Shankaranarayanan, P., Schewe, T., Nigam, S., 2001.
Evidence for the presence of phospholipid hydroperoxide glu-
tathione peroxidase in human platelets: implications for its involve-
ment in the regulatory network of the 12-lipoxygenase pathway of
arachidonic acid metabolism. Biochem. J. 353, 91–100.
Wurzenberger, M., Grosch, W., 1984a. Origin of the oxygen in the
products of the enzymatic cleavage reaction of linoleic acid to 1-
octen-3-ol and 10-oxo-trans-8-decenoic acid in mushrooms (Psal-
liota bispora). Biochim. Biophys. Acta 794, 18–24.
Kajiwara, T., Odake, Y., Hatanaka, A., 1975. Synthesis of 3Z-nonenal
and 3Z, 6Z-nonadienal. Agric. Biol. Chem. 39, 1617–1621.
¨
Lehmann, W.D., Stephan, M., Furstenberger, G., 1992. Profiling
assay for lipoxygenase products of linoleic and arachidonic acid by
gas chromatography-mass spectrometry. Anal. Biochem. 204, 158–170.
Martini, D., Iacazio, G., Perrand, D., Buono, G., Triantaphylides, C.,
1994. Optimization of large scale preparation of 13-(S)-hydro-
peroxy-9Z,11E-octadecadienoic acid using soybean lipoxygenase.
Application to the chemoenzymatic synthesis of (+)-coriolic acid.
Biocatalysis 11, 47–63.
Wurzenberger, M., Grosch, W., 1984b. The formation of 1-octen-3-ol
from the 10-hydroperoxide isomer of linoleic acid by a hydroper-
oxide lyase in mushrooms (Psalliota bispora). Biochim. Biophys.
Acta 794, 25–30.
Matsui, K., Ujita, C., Fugimoto, S., Wilkinson, J., Hiatt, B., Knauf,
V., Kajiwara, T., Feussner, I., 2000. Fatty acid 9- and 13-hydro-
peroxide lyases from cucumber. FEBS Lett. 481, 183–188.
Matsui, K., 1998. Properties and structures of fatty acid hydroper-
oxide lyase. Belg. J. Bot. 131, 50–62.
Wurzenberger, M., Grosch, W., 1986. Enzymic oxidation of linolenic
acid to 1, Z-5-octadien-3-ol, Z-2,Z-5-octadien-1-ol and 10-oxo-E-8-
decenoic acid by a protein fraction from mushrooms (Psalliota bis-
pora). Lipids 21, 261–266.
Noordermeer, M.A., Veldink, G.A., Vliegenthart, J.F.G., 2001. Fatty
acid hydroperoxide lyase: a plant cytrochrom P450 enzyme involved
in wound healing and pest resistance. Chembiochem 2, 494–504.
Weitzel, F., Wendel, A., 1993. Selenoenzymes regulate the activity of
leukocyte 5-lipoxygenase via the peroxide tone. J. Biol. Chem. 268,
6288–6292.