Identification and functional analyses of two cDNAs that encode fatty acid 9-/13-hydroperoxide lyase (CYP74C) in rice

Article abstract of DOI:10.1271/bbb.69.1545

Fatty acid hydroperoxide lyase (HPL), a member of cytochrome P450 (CYP74), produces aldehydes and oxoacids involved in plant defensive reactions. In monocots, HPL that cleaves 13-hydroperoxides of fatty acids has been reported, but HPL that cleaves 9-hydroperoxides is still unknown. To find this type of HPL, in silico screening of candidate cDNA clones and subsequent functional analyses of recombinant proteins were performed. We found that AK105964 and AK107161 (Genbank accession numbers), cDNAs previously annotated as allene oxide synthase (AOS) in rice, are distinctively grouped from AOS and 13-HPL. Recombinant proteins of these cDNAs produced in Escherichia. coli cleaved both 9- and 13-hydroperoxide of linoleic and linolenic into aldehydes, while having only a trace level of AOS activity and no divinyl ether synthase activity. Hence we designated AK105964 and AK107161 OsHPL1 and OsHPL2 respectively. They are the first CYP74C family cDNAs to be found in monocots.

Full text of DOI:10.1271/bbb.69.1545

Biosci. Biotechnol. Biochem., 69 (8), 1545–1554, 2005
Identification and Functional Analyses of Two cDNAs
That Encode Fatty Acid 9-/13-Hydroperoxide Lyase (CYP74C) in Rice
y
Hisao KURODA, Toshiyuki OSHIMA, Hirotaka KANEDA, and Masachika TAKASHIO
Frontier Laboratories of Value Creation, SAPPORO BREWERIES LTD.,
1
0 Okatohme, Yaizu, Shizuoka 425-0013, Japan
Received March 4, 2005; Accepted June 2, 2005
Fatty acid hydroperoxide lyase (HPL), a member of
cytochrome P450 (CYP74), produces aldehydes and oxo-
acids involved in plant defensive reactions. In monocots,
HPL that cleaves 13-hydroperoxides of fatty acids has
been reported, but HPL that cleaves 9-hydroperoxides
is still unknown. To find this type of HPL, in silico
screening of candidate cDNA clones and subsequent
functional analyses of recombinant proteins were per-
formed. We found that AK105964 and AK107161
(HPL, CYP74B,C); and colneleic and colnelenic acid as
plant defensive compounds produced by divinyl ether
6
,7)
synthase (DES, CYP74D). Enzymes belonging to the
CYP74 family are closely related in protein primary
structure and enzymatic function, where they share
common substrate and reaction intermediate called
epoxy allylic carbocation (Fig. 1).
HPLs, widely found in the plant kingdom, are well
characterized in dicots and are divided into two groups
by their substrate specificities (Fig. 1), viz., 13-HPL
(CYP74B) and 9-/13-HPL (CYP74C). 13-HPL prefer-
entially cleaves (9Z,11E,13S)-13-hydroperoxy-9,11-
octadecadienoic acid (13-HPOD, 13-hydroperoxide of
linoleic acid) and (9Z,11E,13S,15Z)-13-hydroperoxy-
9,11,15-octadecatrienoic acid (13-HPOT, 13-hydroper-
oxide of linolenic acid) into hexanal and (3Z)-hexenal
(Genbank accession numbers), cDNAs previously anno-
tated as allene oxide synthase (AOS) in rice, are
distinctively grouped from AOS and 13-HPL. Recombi-
nant proteins of these cDNAs produced in Escherichia.
coli cleaved both 9- and 13-hydroperoxide of linoleic and
linolenic into aldehydes, while having only a trace level
of AOS activity and no divinyl ether synthase activity.
Hence we designated AK105964 and AK107161 OsHPL1
and OsHPL2 respectively. They are the first CYP74C
family cDNAs to be found in monocots.
8
,9)
respectively,
while 9-/13-HPL has both 13-HPL
activity and activity toward (9S,10E,12Z)-9-hydroper-
oxy-10,12-octadecadienoic acid (9-HPOD, 9-hydroper-
oxide of linoleic acid) and (9S,10E,12Z,15Z)-9-hydro-
peroxy-10,12,15-octadecatrienoic acid (9-HPOT, 9-
Key words: fatty acid hydroperoxide lyase; allene oxide
synthase; CYP74; Oryza sativa; cytochrome
P450
1
0)
hydroperoxide of linolenic acid). 9-/13-HPL cleaves
9-HPOD into (3Z)-nonenal and 9-oxo-nonanoic acid,
and cleaves 9-HPOT into (3Z,6Z)-nonadienal and 9-oxo-
nonanoic acid. (3Z)-Hexenal, (3Z)-nonenal and (3Z,6Z)-
nonadienal are enzymatically or spontaneously isomer-
ized into (2E)-hexenal, (2E)-nonenal and (2E,6Z)-non-
adienal respectively. In monocots, only 13-HPL is
Phytooxylipins, which are synthesized from linoleic
and linolenic acids by cascade reactions through lip-
oxygenase, cytochrome P450s, cyclooxygnease, etc., are
thought to play an important role in plant defensive
1
)
9)
reactions and plant development. CYP74 is a cyto-
chrome P450 that transforms fatty acid hydroperoxides
reported in barley, and no presence of 9-/13-HPL
has been reported.
2
)
into various derivatives. CYP74 enzymes are unique in
that they do not require molecular oxygen or NAD(P)H-
dependent cytochrome P450-reductase for their cataly-
ses. Phytooxylipins produced by CYP74 include 12,13-
epoxy-octadecatrienoic acid (allene oxide), a first
intermediate for the plant hormone jasmonic acid,
produced by allene oxide synthase (AOS, CYP74A);3–5)
aldehydes and oxo-acids related to plant defensive
reactions produced by fatty acid hydroperoxide lyases
Aldehydes produced by HPL function as defensive
1
1,12)
compounds in some herbivores.
They are not only
important for plant defensive reactions, but are also
known as negative and positive flavors that determine
1
3–16)
the quality of food and beverages.
brewing, (2E)-nonenal is considered the major contrib-
For example, in
17–19)
utor to the cardboard flavor that arises in aged beer.
(2E)-Nonenal is also one of the off-flavors of aged
milled rice. During investigation of the formation of
2
0)
y
To whom correspondence should be addressed. Fax: +81-54-629-3144; E-mail: Hisao.Kuroda@sapporobeer.co.jp
Abbreviations: AOS, allene oxide synthase; DES, divinyl ether synthase; HPL, fatty acid hydroperoxide lyase; 9-HPOD, (9S,10E,12Z)-9-
hydroperoxy-10,12-octadecadienoic acid; 13-HPOD, (9Z,11E,13S)-13-hydroperoxy-9,11-octadecadienoic acid; 9-HPOT, (9S,10E,12Z,15Z)-9-hydro-
peroxy-10,12,15-octadecatrienoic acid; 13-HPOT, (9Z,11E,13S,15Z)-13-hydroperoxy-9,11,15-octadecatrienoic acid

1546
H. KURODA et al.
Fig. 1. Similarities among Reaction Mechanisms of the Production of Allene Oxide, Aldehydes and Oxo-Acids, and Divinyl Ether by CYP74.
AOS, DES, and HPL share the intermediate epoxy allylic carbocaction formed from the acyl hydroperoxide (9-, 13-HPOD, 9-, 13-HPOT).
AOS and DES deprotonate the carbocation at different positions, which leads to allene oxide and divinyl ether (derived from 9-HPOT and 9-
HPOT) respectively. HPL catalyzes rearrangement and subsequent fragmentation, which leads to the formation of aldehydes and oxo-acids.
Allene oxide derived from 13-HPOT is the precursor of jasmonic acid. Allene oxide is unstable, and without downstream enzymes it
spontaneously transforms into ꢀ- and ꢁ-ketol. R1 and R2 represent C4C7 and (CH2)6COOH (13-HPOT), C4C9 and (CH2)6COOH (13-HPOD),
(CH2)6COOH and C4C7 (9-HPOT), and (CH2)6COOH and C4C9 (9-HPOD) respectively. Symbols: AOS, allene oxide synthase; DES, divinyl
ether synthase; HPL, fatty acid hydroperoxide lyase; 13-HPL, fatty acid 13-hydroperoxide lyase; 9-/13-HPL, fatty acid 9-/13-hydroperoxide
lyase.
(
(
2E)-nonenal in the brewing process, we found that
2E)-nonenal was produced by barley lipoxygenase-1
tively, are the first CYP74C members to be found in
monocots.
2
1)
and fatty acid 9-hydroperoxide lyase-like activity. The
fatty acid 9-hydroperoxide lyase-like activity is likely to
be an enzyme because it is sensitive to proteinase-K
treatment and the activity is terminated by boiling. This
finding led us to search for HPL that can cleave 9-HPOD
in monocots including rice; however, despite intensive
analysis of full-length cDNA in rice, cDNAs that encode
Materials and Methods
Chemicals. (2E)-Nonenal, hexanal, 13-HPOD, 13-
HPOT and 2-hydroxydecanoic acid were purchased
from Wako Pure Chemical Industries (Osaka, Japan). 9-
HPOD was produced through the reaction of recombi-
2
2)
9
-/13-HPL (CYP 74C) have never been found. We
nant barley LOX-1 and linoleic acid, and purified with
a silica-gel column (silica-gel 60, Merck, U.S.A.). (3Z)-
Nonenal, colneleic acid, 9-HPOT, 13-HPOT, (9Z,15Z)-
13-hydroxy-12-oxo-9,15-octadecadienoic acid, (10E,15Z)-
9-hydroxy-12-oxo-10,15-octadecadienoic acid, (10E)-9-
hydroxy-12-oxo-10-octadecenoic acid and 9-oxo-non-
anoic acid were purchased from Larodan Fine Chem-
icals (Malm o¨ , Sweden). (2E,6Z)-nonadienal was pur-
chased from Sigma-Aldrich (St. Louis, MO).
thought that they have not been identified because of the
similarity of their structure to AOS, 13-HPL and DES,
and that they are incorrectly annotated. Hence we
attempted in silico screening of candidate cDNA clones
and tested their recombinant proteins as 9-/13-HPL.
Here we show that AK105964 and AK107161, previ-
ously designated AOS, are 9-/13-HPL. AK105964 and
AK107161 designated OsHPL1 and OsHPL2 respec-

Fatty Acid 9-/13-Hydroperoxide Lyase (CYP74C) in Rice
1547
In silico screening, molecular cloning, and E. coli
expression of OsHPL1 and OsHPL2 cDNAs. Candidate
cDNAs, which have high homology to functionally
assigned barley AOS1 and AOS2 (designated HvAOS1
and HvAOS2 in this paper; Genbank accession
protein assay (Bio-Rad, Hercules, CA, U.S.A.). An
aliquot (18 mg) of E. coli lysate was subjected to Ni-
NTA superflow (bed volume 0.25 ml, Qiagen, Hilden,
Germany). After washing the resin with 10 ml of 10 mM
Tris–HCl (pH 7.2), 0.1% Tween 20, 1 mM ꢂ-mercap-
toethanol, 10 mM imidazole, and 0.3 M NaCl, His-tagged
protein was eluted with 0.75 ml of 10 mM Tris–HCl
(pH 7.2), 0.1% Tween 20, 1 mM ꢂ-mercaptoethanol,
250 mM imidazole, 0.3 M NaCl. The eluate was concen-
trated with Centricon 5000 (Waters, Massachusetts,
U.S.A.) at a final volume of 100 ml to obtain an affinity-
purified fraction. An aliquot of the affinity-purified
fraction (2 ml) was analyzed by SDS–PAGE (5–20%
gradient gel).
2
3)
nos. AJ250864 and AJ251304 respectively), barley
3-HPL (designated Hv13-HPL in this paper; Genbank
1
accession no. AJ318870) and rice AOS (OsAOS;
9
)
2
4)
Genbank accession no. AY055775) were extracted
from the Knowledge-based Oryza Molecular Biological
Encyclopedia (KOME) database and the Expert Protein
Analysis System (Swiss-Prot and TrEMBL), using the
protein–protein BLAST program (threshold, E-value =
ꢀ10
e
). Phylogenetical analysis was conducted with the
DNASIS Pro program (Hitachi Software Engineering,
Tokyo, Japan).
PolyA RNA was isolated from etiolated shoots from
Enzyme assays and product analysis. GC–MS analy-
sis of aldehydes produced by OsHPL1 and OsHPL2
proteins was as follows: E. coli lysate (100 mg) was
diluted to 1 ml of 0.1 M Tris–HCl (pH 7.0) in a GC vial
and mixed with either 9-, 13-HPOD or 9-, 13-HPOT at a
final concentration of 0.1 mM. After capping, a 100 mm
membrane-thick polydimethylsiloxane solid phase mi-
cro extraction fiber probe (Supelco, Bellefonte, PA) was
inserted into the headspace of the vial, and further
þ
rice (Oryza sativa, cv. Nipponbare) using Tripure
isolating reagent (Roche Diagnostics, Tokyo, Japan)
and Oligotex dT30 super (Takara, Kyoto, Japan). Single-
stranded cDNA was synthesized with Superscript II and
oligo(dT)20 primer (Invitrogen, Carlsbad, CA). cDNAs
corresponding to the longest ORF of AK105964 and
AK107161 were produced by polymerase chain
0
ꢁ
reaction using primers 5 -gcggatccatggcgcgccgccgc-
0
incubated at 40 C for 5 min. The probe was then
introduced into the injection port and thermally des-
0
0
gagccaa-3 and 5 -gcgaattcccgcaccaacactcgccgctc-3 for
0
ꢁ
AK105964, and primers 5 -gccggatccatggcgccaccgc-
0
orbed for 5 min at 260 C onto a DB-1 column (liquid
0
0
cagt-3 and 5 -gccgaattccgcaattaaaacatccaccaacct-3 for
AK107161, with KOD-Plus- DNA polymerase (Toyobo,
phase = dimethylpolysiloxane, 30 m ꢂ 0:25 mm, film
thickness = 1 mm, J&W Scientific, CA, U.S.A.). MS
analysis was performed with the Hewlett Packard
HP6890/MSD system in scan mode. Products were
identified by mass-spectrum analysis and the coinci-
dence of retention time with authentic compound except
for (3Z,6Z)-nonadienal.
ꢁ
Tokyo, Japan). The PCR program was 94 C for 2 min,
ꢁ
ꢁ
4
0 cycles of 94 C for 15 sec and 68 C for 4 min, and
ꢁ
6
8 C for 10 min. Amplified fragments were blunt-ended
and cloned into the HincII site of pUC118, and
subsequently subcloned into the BamHI–EcoRI site of
E. coli expression vector, pRSETA (Invitrogen). These
plasmids (designated pRSETA-OsHPL1 and pRSETA-
OsHPL2 for AK105964 and AK107161 respectively)
are designed to express a recombinant protein with a tag
peptide containing a vector sequence and a hexa-
histidine metal binding domain for affinity-purification
Production of 9-oxo-nonanoic acid from 9-HPOD and
9-HPOT, colneleic acid from 9-HPOD, ꢀ- and ꢁ-ketol
from 9-, 13-HPOD and 9-, 13-HPOT by E. coli lysate
of OsHPL1 and OsHPL2 was measured by LC–MS as
follows. E. coli lysate (10 mg) was diluted to 0.1 ml of
0.1 M Tris–HCl (pH 7.0) and incubated with either 9-,
13-HPOD or 9-, 13-HPOD at a final concentration of
(His-tagged protein) consisting of MRGSHHHHHH-
ꢁ
GMASMTGGQQMGRDLYDDDDKDRWGS (36 resi-
dues). After confirming that the DNA sequences were
identical to the data present in the KOME database,
expression vectors (pRSETA, pRSETA-OsHPL1 and
pRSETA-OsHPL2) were transformed into BL21 (DE3)
pLysS. These strains were cultured separately in 100 ml
0.1 mM at 30 C for 20 min. The reaction was terminated
by adding 0.1 ml of ethanol containing 10 ppm of 2-
hydroxydecanoic acid as an internal standard. The
sample was filtered with a Centricon 5000 (Waters),
and the eluate (5 ml) was subjected to HPLC (Hewlett
Packard series 1100, Agilent Technologies, CA, U.S.A.)
equipped with C18 reverse-phase HPLC column (Asym-
ꢁ
of L-broth at 16 C until the cell density reached
ꢁ
ꢁ
OD600 ¼ 0:7, and further cultured at 16 C for 16 h after
metry, 3.5 mm, 2:1 ꢂ 150 mm, Waters) at 50 C. The
the addition of isopropyl-ꢂ-thiogalactopyranoside at a
final concentration of 2 mM. Bacterial cells were
recovered by centrifugation and suspended in 10 ml of
mobile phase consisted of solvent A (0.5% acetic acid)
and solvent B (acetonitrile). The HPLC run program
was a linear gradient of 100% A to 100% B (0–10 min),
following 100% B (10–16 min); the flow rate was
0.3 ml/min. MS analysis was performed by Micromass
ZQ (Waters) in negative electrospray ionization mode
(capillary voltage ꢀ3:2 kV, source and desolvation
1
0 mM Tris–HCl (pH 7.2), 0.1% Tween 20, and 1 mM ꢂ-
mercaptoethanol. After freezing and thawing, the sample
was incubated with 100 U of DNase I (Nippon Gene,
ꢁ
Tokyo, Japan) at 25 C for 10 min, and then centrifuged
at 10;000 ꢂ g to obtain E. coli lysate. The protein
ꢁ
ꢁ
temperature 120 C and 350 C respectively). Products
were measured in select-ion mode using 2-hydroxyde-
content of E. coli lysate was determined by Bio-Rad

1548
H. KURODA et al.
Fig. 2. Phylogenetical Analysis (unrooted tree) of Candidate cDNAs in Rice Having High Homology to Rice AOS, Barley AOS and Barley 13-
HPL.
The databases used were the Knowledge-based Oryza Molecular Biological Encyclopedia (KOME) database (http://cdna01.dna.affrc.go.jp/
cDNA/) and the Expert Protein Analysis System (Swiss-Prot and TrEMBL, http://us.expasy.org/). cDNAs are designated as gene names and/or
Genbank accession numbers.
canoic acid as an internal standard. The pseudo-mo-
lecular ions monitored were m=z of 187.1 (2-hydroxy-
decanoic acid), 171.1 (9-oxo-nonanoic acid), 309.2 (ꢀ-
and ꢁ-ketol from 9- and 13-HPOT), 311.2 (ꢀ- and ꢁ-
ketol from 9- and 13-HPOD), and 293.2 (colneleic acid).
We also monitored m=z of 211.2 for detection of (9Z)-
nated AK105964 and AK107161, which were distinc-
tively grouped from 13-HPL and AOS (Fig. 2).
AK105964 and AK107161 were previously designated
OsAOS4 and OsAOS3 respectively, but the catalytic
activities of the recombinant proteins of these cDNAs
2
5)
were not clarified. AK105964 and AK107161 were
also annotated as putative AOS on the KOME database,
but we thought that they might have a catalytic activity
of 9-/13-HPL. To confirm this, we cloned cDNAs
corresponding to the longest open reading frame of
1
2-oxo-9-dodecenoic acid from 13-HPOD and 13-
HPOT, though the amount was not determined due to
the unavailability of authentic compound.
The substrate specificities of OsHPL1 and OsHPL2
were assayed spectrophotometrically by monitoring
decreases in the substrates as follows: 9-, 13-HPOD or
þ
AK105964 and AK107161 by RT-PCR from polyA -
RNA extracted from etiolated shoots. The nucleotide
sequences of cloned AK105964 and AK107161 were
identical to the data presented on the KOME database.
For simplicity, from this point we temporarily designate
AK105964 and AK107161 OsHPL1 and OsHPL2
respectively. The deduced amino acids of OsHPL1 and
OsHPL2 corresponding to the longest open reading
frames are shown in Fig. 3. They have calculated
molecular masses of 55.3 kDa and 54.4 kDa, comprising
of 510 and 500 amino acid residues respectively. The
longest open reading frames of OsHPL1 and OsHPL2
were subcloned to pRSETA expression vector, and we
designated them pRSETA-OsHPL1 and pRSETA-
OsHPL2 respectively. These vectors are designed to
express recombinant protein of either OsHPL1 or
OsHPL2 protein with a N-terminal tag peptide (MRGS-
HHHHHHGMASMTGGQQMGRDLYDDDDKDRW-
GS, 36 residues), which contains amino acid residues
derived from the vector sequence and hexa-histidine
metal binding domain for affinity-purification (the
His-tagged protein).
9
-, 13-HPOD was diluted to 1 ml of 0.1 M Tris–HCl
(
pH 7.0) in a glass cell at a final concentration of 40 mM.
An aliquot of the affinity-purified fraction (2 ml) was
mixed in, and the absorbance at 234 nm was monitored
at room temperature. Activity was determined as the rate
of consumption of conjugated diene chromophore (" ¼
2
5;000).
Results
In silico screening, molecular cloning, and E. coli
expression of OsHPL1 and OsHPL2 cDNAs
In silico screening for rice proteins that have a high
homology to functionally assigned barley AOS1 and
AOS2 (designated HvAOS1 and HvAOS2 in this paper;
Genbank accession no. AJ250864 and AJ251304 re-
spectively),2 barley 13-HPL (designated as Hv13-HPL
in this paper; Genbank accession No., AJ318870) and
rice AOS (designated as OsAOS in this paper; Genbank
accession No., AY055775), following phylogenetical
analysis revealed the presence of two cDNAs, desig-
3)
9
)
2
4)

Fatty Acid 9-/13-Hydroperoxide Lyase (CYP74C) in Rice
1549
Fig. 3. Comparison of the Protein Sequences of OsHPL1, OsHPL2, Barley AOS, and Rice AOS.
The amino acid sequences of OsHPL1, OsHPL2, OsAOS, OsAOS1, HvAOS1 and HvAOS2 were aligned using the ClastalW 1.8 program
(
http://clustalw.genome.jp/). Black boxes indicate amino acid residues that are conserved between all sequences. Subfamily-specific amino acid
residues are indicated with an asterisk. The I-helix GXXX(F/L), EXLR motif and PXVXNKQCPG for heme-binding domain are double-
underlined. The PPGP tetrapeptide motif for catalytic stability of the cytochrome P450 family is underlined by black bars. The fusion proteins of
OsHPL1 and OsHPL2 contain 36 residues of a N-terminal tag-sequence (MRGSHHHHHHGMASMTGGQQMGRDLYDDDDKDRWGS) in
front of the methionine of the indicated sequence.

1550
H. KURODA et al.
Fig. 4. GC–MS Analysis of Aldehydes Produced by the Incubation of E. coli Lysate of pRSETA (vector control) and pRSETA-OsHPL1 with Either
9-, 13-HPOD or 9-, 13-HPOD.
Total ion chromatographs are presented. The peaks indicated by arrows were identified as respective compounds by coincidence of the
retention time of authentic compounds and analysis of mass spectra, except for (3Z,6Z)-nonadienal. Double-headed arrows represent a scale for
signal intensity. Similar results were obtained for pRSETA-OsHPL2.
þ
þ
GC–MS and LC–MS analysis of the products formed
by OsHPL1 and OsHPL2 with various fatty acid
hydroperoxides
of 123 (M –CH3), 109 (M –CHO), 95, 79, 67, 55
þ
(C4H7 ), was an isomer of nonadienal, putatively
identified as (3Z,6Z)-nonadienal. Incubation of 13-
HPOD with the lysate of pRSETA-OsHPL1 resulted in
the production of hexanal (m=z ¼ 82, 72, 56, 44).
Incubation of 9-HPOD, 9-HPOT and 13-HPOD with the
lysate of pRSETA-OsHPL1 resulted in the formation of
other trace peaks, which were also observed in pRSETA
(insert-free vector control), and were from the host or
non-enzymatic transformation of substrates. Incubation
of 13-HPOT with the lysate of pRSETA-OsHPL1, but
not with that of pRSETA, resulted in the formation of a
peak identified as (3Z)-hexenal (m=z ¼ 98, 83, 69, 55,
41) and a trace peak of (2E)-hexenal (m=z ¼ 98, 83, 69,
55, 41). The other signals appeared at retention times of
8–18 min in pRSETA-OsHPL1, which signals were also
observed for pRSETA, and were again from the host or
non-enzymatic transformation of 13-HPOT.
To identify the catalytic property of OsHPL1 and
OsHPL2 as HPL, we firstly performed a qualitative
analysis with GC–MS, whether OsHPL1 and OsHPL2
produce aldehydes with incubation of various fatty acid
hydroperoxides. Since similar results were obtained for
OsHPL1 and OsHPL2, here we show only the result
obtained for OsHPL1. Figure 4 summarizes the total-ion
chromatographs of the aldehydes produced by the
incubation of E. coli lysate of pRSETA-OsHPL1 with
9
-, 13-HPOD, or 9-, 13-HPOT. E. coli lysate of
pRSETA-OsHPL1 cleaved 9-HPOD into a main product
of (3Z)-nonenal (m=z ¼ 122, 111, 96, 84, 69, 55, 41) and
a minor product of (2E)-nonenal (m=z ¼ 122, 111, 96,
8
3, 70, 55, 41). Part of the production of (2E)-nonenal
was due to the spontaneous isomerization of (3Z)-
nonenal, as was confirmed by GC–MS analysis of an
authentic compound. Incubation of 9-HPOT with the
lysate of pRSETA-OsHPL1 resulted in the formation of
a major product with a minor product of (2E,6Z)-
nonadienal [m=z ¼ 138, 123, 109, 70, 41]. The major
peak, which had a parent ion of 138 and fragment ions
Secondly, to determine the main catalytic activity of
OsHPL1 and OsHPL2 as HPL, AOS or DES, the metabo-
lites produced from either 9-, 13-HPOD, or 9-, 13-HPOT
were analyzed by LC–MS. To test for OsHPL1 or
OsHPL2 as DES activity, E. coli lysates of pRSETA-
OsHPL1 and pRSETA-OsHPL2 were incubated with 9-

Fatty Acid 9-/13-Hydroperoxide Lyase (CYP74C) in Rice
1551
contrast, OsHPL1 and OsHPL2 efficiently transformed
-HPOD and 9-HPOT into 9-oxo-nonanoic acid
Fig. 5A). We monitored m=z of 171 to quantify the
9
(
production of 9-oxo-nonanoic acid, and found that the
transformation rate was 72–80% (Table 1). When 13-
HPOD or 13-HPOT was incubated with E. coli lysate of
pRSETA-OsHPL1 and pRSETA-OsHPL2, we detected a
single peak at m=z of 211 at a retention time of 10.5 min,
which corresponds to (9Z)-12-oxo-9-dodecenoic acid
(data not shown). Although we could not quantify the
amount of (9Z)-12-oxo-9-dodecenoic acid due to the
unavailability of the authentic compound, the signal
intensity was similar to that of 9-oxo-nonanoic acid,
suggesting that the same level of (9Z)-12-oxo-9-dodec-
enoic acid was produced as of 9-oxo-nonanoic acid.
Correctively, the results of GC–MS and LC–MS analyses
indicate that the major activity of OsHPL1 and OsHPL2
is HPL. Thus, OsHPL1 (AK105964) and OsHPL2
(AK107161) cDNAs previously annotated as allene
oxide synthase in rice have 9-/13-fatty acid hydroper-
oxide lyase activity. OsHPL1 and OsHPL2 are the first
CYP74C family enzymes to appear in monocots.
Fig. 5. LC–MS Analysis of the Products Generated by Incubation of
-HPOD and E. coli Lysates of pRSETA, pRSETA-OsHPL1 or
pRSETA-OsHPL2.
9
Substrate specificity of OsHPL1 and OsHPL2
E. coli lysate was incubated with 9-HPOD at a final concentration
of 0.1 mM, and the reaction was terminated by adding an equal
volume of ethanol containing an internal standard. Colneleic acid
was detected by monitoring m=z of 293.2. The double-headed arrow
represents a scale for signal intensity.
Finally, to compare the catalytic activities of OsHPL1
and OsHPL2, we attempted to purify recombinant
proteins from E. coli lysates of pRSETA-OsHPL1 and
pRSETA-OsHPL2. The condition of recombinant pro-
ꢁ
ꢁ
tein was pre-culture at 16 C and induction at 16 C. A
preliminary experiment revealed that no visible re-
combinant protein was observable in the affinity-purified
HPOD. We did not detect any colneleic acid, a divinyl
ether converted by DES from 9-HPOD, indicating that
neither OsHPL1 nor OsHPL2 has DES activity (Fig. 5).
In plant cells, AOS catalyzes the transformation of 13-
HPOT into (9Z,11E,15Z,12R,13S)-12,13-epoxy-9,11,15-
octadecarienoic acid (allene oxide), which is subsequent-
ly metabolized into cis-(+)-12-oxo-phytodienoic acid by
allene oxide cyclase, leading to jasmonate metabolism.
In the absence of allene oxide cyclase, this allene oxide is
chemically hydrdolyzed into (9Z,15Z)-13-hydroxy-12-
oxo-9,15-octadecedienoic acid (ꢀ-ketol, about 80%),
ꢁ
sample purified from E. coli pre-cultured at 37 C
ꢁ
and induction at 16 C by SDS–PAGE analysis. Both
pRSETA-OsHPL1 and pRSETA-OsHPL2, but not
pRSETA, produced a single protein with an apparent
molecular weight of 54 kDa on SDS–PAGE, which was
good agreement with the calculated molecular weights
of 59.6 kDa and 58.7 kDa of the His-tagged fusion
protein of OsHPL1 and OsHPL2 respectively. However,
the expression levels of fusion proteins of OsHPL1 and
OsHPL2 were very low, so that affinity-purified frac-
tions contained co-purified host proteins that bound to
the Ni-NTA affinity column.
(
(
10E,15Z)-9-hydroxy-12-oxo-10,15-octadecadienoic acid
ꢁ-ketol, about 10%) and racemic cis-12-oxo-phytodi-
enoic acid. 13-HPOD is also metabolized by AOS, and in
addition to 13-hydroperoxides, it has been shown that
two barley AOSs (HvAOS1 and HvAOS2) are able to
Using these affinity-purified fractions, we compared
the substrate specificity of OsHPL1 and OsHPL2. The
relative activity toward 9-HPOT, 9-HPOD, 13-HPOT
and 13-HPOD was 100:66:20:61 for OsHPL1, and
100:93:85:59 for OsHPL2, respectively. This was in
sharp contrast to 13-HPL (CYP74B) found in barley that
specifically metabolizes 13-hydroperoxides of linoleic
2
3)
catalyze 9-HPOD and 9-HPOT. To test for OsHPL1
and OsHPL2 as AOS, we incubated 9-, 13-HPOD, and
9
-, 13-HPOT with E. coli lysate of pRSETA-OsHPL1
and pRSETA-OsHPL2, and monitored for m=z of 309 (ꢀ-
and ꢁ-ketol from 9- or 13-HPOT) and m=z of 311 (ꢀ- and
9
)
ꢁ
-ketol from 9- or 13-HPOD). Incubation of 13-HPOT
and linolenic acid. Both OsHPL1 and OsHPL2 showed
highest activity toward 9-HPOT, and less activity
toward 9-HPOD and 13-HPOD. The remarkable differ-
ence in substrate specificity between OsHPL1 and
OsHPL2 was the activity toward 13-HPOT. OsHPL2
showed similar levels of activity toward 9-HPOT and
13-HPOT, while the activity of OsHPL1 toward 13-
HPOT was only 20% of that toward 9-HPOT.
resulted in the formation of (9Z,15Z)-13-hydroxy-12-
oxo-9,15-octadecedienoic acid (ꢀ-ketol, Fig. 6B), but the
transformation rates observed were only 1.3% and 2.5%
for OsHPL1 and OsHPL2 respectively (Table 1). But, we
did not detect any ꢀ- or ꢁ-ketol by incubation of 9-, 13-
HPOD or 9-HPOT (Fig. 6B, C). This indicates that AOS
is not the major activity of OsHPL1 or OsHPL2. In

1552
H. KURODA et al.
Fig. 6. LC–MS Analysis of the Products Generated by Incubation of Various Fatty Acid Hydroperoxides and E. coli Lysates of pRSETA, pRSETA-
OsHPL1 and pRSETA-OsHPL2.
Each E. coli lysate was incubated with the indicated fatty acid hydroperoxide at a final concentration of 0.1 mM, and the reaction was
terminated by adding an equal volume of ethanol containing an internal standard. Production of 9-oxo-nonanoic acid from 9-HPOD and 9-HPOT
(A), ꢀ- and ꢁ-ketol from 9-HPOT and 13-HPOT (B), and ꢀ- and ꢁ-ketol from 9-HPOD or 13-HPOD (C) was detected by monitoring m=z of
171.1, 309.2, and 311.2, respectively. Double-headed arrows represent scales for signal intensity. Symbols: ꢀ-ketol, (9Z,15Z)-13-hydroxy-12-
oxo-9,15-octadecadienoic acid; ꢁ-ketol, (10E,15Z)-9-hydroxy-12-oxo-10,15-octadecadienoic acid.
Discussion
Table 1. Quantification of the Products Formed by OsHPL1 or
OsHPL2 with Various Fatty Acid Hydroperoxides
We have shown that AK105964 and AK107161,
2
5)
Products (nmol)
-oxo-nonanoic acid ꢀ-ketol ꢁ-ketol
cDNAs previously annotated as AOS, had a catalytic
activity of 9-/13-HPL, but had no DES activity and only
a trace level of AOS activity. We designated AK105964
and AK107161 OsHPL1 and OsHPL2 respectively, and
these are the first members of CYP74C found in
monocots. As previously described, to date seven
cDNAs have been annotated as AOS in rice. This study
indicates that two of them are 9-/13-HPL. AOS is a key
enzyme for the biosynthesis of jasmonates, important
plant hormones that regulate the expression of stress
proteins and pathogen-defense related genes, growth,
pollen maturation and release, whereas 9-/13-HPL
produces completely different metabolites, aldehydes
and oxo-acids, for plant-microbe or plant-insect inter-
actions. From this point of view, this study clarifies the
Enzyme Substrates
9
OsHPL1
9-HPOD
3-HPOD
7.2
—
ND
ND
ND
0.25
ND
ND
ND
0.13
ND
ND
ND
ND
ND
ND
ND
ND
1
1
1
1
9-HPOT
7.2
—
3-HPOT
9-HPOD
3-HPOD
OsHPL2
8.0
—
9-HPOT
3-HPOT
7.7
—
Ten nmol of fatty acid hydroperoxides was incubated with E. coli lysates.
Values are the means of duplicate measurements.
—, not tested.
ND, not detected.

Fatty Acid 9-/13-Hydroperoxide Lyase (CYP74C) in Rice
1553
classification of the members of the CYP74 family in
rice, and thus would help in understanding the physio-
logical roles of AOS and 9-/13-HPL genes in rice.
As shown in Fig. 2, phylogenetical analysis indicated
that OsHPL1 and OsHPL2 are more closely related to
CYP74A than to CYP74B, and this might be one of the
reasons for the incorrect annotation of OsHPL1 and
OsHPL2 as AOS (Fig. 1). The primary structures
of 9-/13-HPL and AOS of rice are well resembled.
Comparisons of the amino acid sequences of OsHPL1,
OsHPL2, OsAOS, OsAOS1 (Genbank accession
phytochrome. OsHPL1 is up-regulated by red and far-
red light in seedling shoots. On the other hand, the
2
5)
transcript of OsHPL2 was too low to determine. To
date, there is no evidence that HPL is controlled by
phytochrome, and this sheds a new light on 9-/13-HPL
as related to light regulation of gene expression.
OsHPL1 and OsHPL2 are very closely located at
31.7 cM on the chromosome 2 (locus name, AP004996;
KOME database). From sequence analysis of PAC clone
P0027A02, the coding regions of OsHPL1 and OsHPL2
are located at 71520–73052 bp and 61930–63432 bp
respectively. Between OsHPL1 and OsHPL2, a hypo-
thetical ORF, which codes for a peptide with 111 amino
acid residues predicted at 64835–65170 bp (Protein ID
no. BAD17183.1) has been identified. The regulation
and evolution of OsHPL1 and OsHPL2 gene is for
further work.
2
5)
no. AB116527),
1
HvAOS1 and HvAOS2 revealed
65 positions of conserved amino acids (Fig. 3). These
include the I-helix GXXX(F/L) and EXLR motifs
involved in the O2 binding reported in dicotyledoneous
species,2 PPGP tetrapeptide motif for the catalytic
6)
2
7)
stability of the cytochrome P450 family,
and
2
3)
PXVXNKQCPG for the heme-binding domain. On
the other hand, we observed 34 subfamily-specific
amino acids that might reflect the differences in catalytic
activity between AOS and 9-/13-HPL. In dicots, some
subfamily-specific amino acids were found between
HPL and AOS, and these are proposed for the dif-
ferences in catalytic activity. For example, the nine
amino acid residues upstream of the cysteine involved in
heme binding is always glutamic acid in AOS, while that
In contrast to C aldehydes, the roles of C aldehydes
9
6
in plant defensive reaction are unclear. Identification of
OsHPL1 and OsHPL2 enabled a transgenic approach to
analyze C9 aldehydes in plant defensive reactions. In
addition, since (2E)-nonenal is considered to be the aged
2
0,21)
flavor in cereal products, including rice and beer,
rice germplasms that lack OsHPL1 and/or OsHPL2
might contribute to the development of a new rice
cultivar with improved food quality.
1
0,26)
of HPL is a neutral amino acid.
However, the
corresponding amino acid residues in OsAOS and
OsAOS1 are glutamic acid and alanine respectively,
and in OsHPL1 and OsHPL2 they are glycine and
arginine respectively. Further study is needed to eluci-
date the structure-function relationship that determines
AOS and HPL activity. It is intriguing that OsAOS1,
whose catalytic activity has not yet been reported, is
more closely related to OsHPL1 and OsHPL2. The
catalytic activity of OsAOS1, which is characterized as
the gene for cpm-1 that controls the process of anthesis
Acknowledgments
We thank Koji Takazumi for assistance with GC–MS.
We also thank Kazuhisa Yasui for helpful discussions
and assistance with LC–MS analysis.
References
1) Blee, E., Phytooxylipins and plant defense reactions.
Prog. Lipid Res., 37, 33–72 (1998).
2
)
)
Feussner, I., and Wasternack, C., The lipoxygenase
pathway. Annu. Rev. Plant Biol., 53, 275–297 (2002).
Agrawal, G. K., Rakwal, R., Jwa, N.-S., Han, K.-S., and
Agrawal, V. P., Molecular cloning and mRNA expres-
sion analysis of the first rice jasmonate biosynthetic
pathway gene allene oxide synthase. Plant Physiol.
Biochem., 40, 771–782 (2002).
Song, W.-C., and Brash, A. R., Purification of an allene
oxide synthase and identification of the enzyme as a
cytochrome P-450. Science, 253, 781–784 (1991).
and phytochrome-mediated inhibition of coleoptile
growth,25) is of interest.
3
The primary structures of OsHPL1 and OsHPL2 are
very similar; they share 85% of their amino acid
residues. It has been shown that OsHPL1 and OsHPL2
2
5)
have no chroloplast transit peptide, but the local-
ization of OsHPL1 and OsHPL2 into chloroplast is not
precluded, because immunochemical analysis has re-
vealed that HvAOS1 and HvAOS2 that lack chloroplast
4
)
2
3)
transit peptides are localized within chloroplasts. The
expression levels of OsHPL1 and OsHPL2 proteins in
E. coli showed differently. The level of OsHPL1 was
much lower than that of OsHPL2 (data not shown). This
might reflect the different toxicity or stability in E. coli.
The remarkable difference between OsHPL1 and
OsHPL2 is the catalytic activity toward 13-HPOT, but
the biological meaning of this is unclear. To speculate
on the difference in the physiological roles of OsHPL1
and OsHPL2, it is interesting that Haga et al. found that
the expression of OsHPL1 (described as OsAOS4), but
not of OsHPL2 (described as OsAOS3) is regulated by
5) Song, W.-C., Funk, C. D., and Brash, A. R., Molecular
cloning of an allene oxide synthase: a cytochrome P450
specialized for the mechanism of fatty acid hydroper-
oxides. Proc. Natl. Acad. Sci. U.S.A., 90, 8519–8523
(
1993).
6
7
)
)
Weber, H., Ch e´ telat, A., Caldelari, D., and Farmer, E. E.,
Divinyl ether fatty acid synthesis in late blight-diseased
potato leaves. Plant Cell, 11, 485–493 (1999).
Itoh, A., and Howe, G. A., Molecular cloning of a
divinyl ether synthase. J. Biol. Chem., 276, 3620–3627
(
2001).
8) Matsui, K., Shibutani, M., Hase, T., and Kajiwara, T.,
Bell pepper fruit fatty acid hydroperoxide lyase is a

1554
H. KURODA et al.
cytochrome P450 (CYP74B). FEBS Lett., 394, 21–24
1996).
Koeduka, T., Stumpe, M., Matsui, K., Kajiwara, T., and
Feussner, I., Kinetics of barley FA hydroperoxide lyase
are modulated by salts and detergents. Lipids, 38, 1167–
H., Yoshida, I., Takashio, M., Ito, K., and Takeda, K.,
Brewing performance of lipoxygenase-1-less barley. In
‘‘World Brewing Congress 2004 Proceeding’’, Master
Brewers Association of the Americas and American
Society of Brewing Chemists, CD-ROM (2004).
20) Lam, H. S., and Proctor, A., Milled rice oxidation
volatiles and odor development. J. Food Sci., 68, 2676–
2681 (2002).
(
9
)
1172 (2003).
1
1
0) Matsui, K., Ujita, C., Fujimoto, S., Wilkinson, J., Hiatt,
B., Knauf, V., Kajiwara, T., and Feussner, I., Fatty acid
9- and 13-hydroperoxide lyases from cucumber. FEBS
Lett., 481, 183–188 (2000).
21) Kuroda, H., Furusho, S., Maeba, H., and Takashio, M.,
Characterization of factors involved in the production of
2(E)-nonenal during mashing. Biosci. Biotechnol. Bio-
chem., 67, 691–697 (2003).
22) Kuroda, H., Kobayashi, N., Kaneda, H., Watari, J., and
Takashio, M., Characterization of factors that transform
linoleic acid into di- and trihydroxyoctadecenoic acids in
mash. J. Biosci. Bioeng., 93, 73–77 (2002).
23) Maucher, H., Hause, B., Feussner, I., Ziegler, J., and
Wasternack, C., Allene oxide synthases of barley
(Hordeum vulgare cv. Salome): tissue specific regulation
in seedling development. Plant J., 21, 199–213 (2000).
24) Ha, S., Lee, B., Lee, D., Kuk, Y. I., Lee, A., Han, O., and
Back, K., Molecular characterization of the gene
encoding rice allene oxide synthase and its expression.
Biosci. Biotechnol. Biochem., 66, 2719–2722 (2002).
25) Haga, K., and Iino, M., Phytochrome-mediated
transcriptional up-regulation of ALLENE OXIDE
SYNTHASE in rice seedlings. Plant Cell Physiol., 45,
119–128 (2004).
1) Vancanneyt, G., Sanz, C., Farmaki, T., Paneque, M.,
Ortego, F., Casta n˜ era, P., and S a´ nchez-Serrano, J. J.,
Hydroperoxide lyase depletion in transgenic potato
plants leads to an increase in aphid performance. Proc.
Natl. Acad. Sci. U.S.A., 98, 8139–8144 (2001).
1
1
2) Kessler, A., Halitschke, R., and Baldwin, I. T., Silencing
the jasmonate cascade: induced plant defenses and insect
populations. Science, 305, 665–668 (2004).
3) Urbach, G., and Gordon, M. H., Flavours derived from
fats. In ‘‘Fats in Food Products’’, eds. Moran, D. P. J.,
and Rajah, K. K., Blackie Academic and Professional,
Glasgow, pp. 347–405 (1994).
1
1
4) Hatanaka, A., The biogeneration of green odour by green
leaves. Phytochemistry, 34, 1201–1218 (1993).
5) Palma-Harris, C., McFeeters, R. F., and Fleming, H. P.,
Fresh cucumber flavor in refrigerated pickles: compar-
ison of sensory and instrumental analysis. J. Agric. Food
Chem., 50, 4875–4877 (2002).
1
6) P e´ rez, A. G., Sanz, C., Ol ´ı ass, R., and Ol ´ı ass, J. M.,
Lypoxygenase and hydroperoxide lyase activities in
ripening strawberry fruits. J. Agric. Food Chem., 47,
26) Howe, G. A., Lee, G. I., Itoh, A., Li, L., and DeRocher,
A. E., Cytochrome P450-dependent metabolism of
oxylipins in tomato. Cloning and expression of allene
oxide synthase and fatty acid hydroperoxide lyase. Plant
Physiol., 123, 711–724 (2000).
27) Szczesna-Skorupa, E., Straub, P., and Kemper, B.,
Deletion of a conserved tetrapeptide, PPGP, in P450
2C2 results in loss of enzymatic activity without a
change in its cellular location. Arch. Biochem. Biophys.,
304, 170–175 (1993).
249–253 (1999).
1
1
1
7) Drost, B. W., van den Berg, R., Freijee, F. J. M., van der
Velde, E. G., and Hollemans, M., Flavor stability. J. Am.
Soc. Brew. Chem., 48, 124–131 (1990).
8) Meilgaard, M. C., Flavor chemistry of beer. PartII.
Flavor and threshold of 239 aroma volatiles. MBAA
Tech. Q., 12, 151–168 (1975).
9) Hirota, N., Kuroda, H., Takoi, K., Kaneko, T., Kaneda,