Molecular cloning of CYP76B9, a cytochrome P450 from Petunia hybrida, catalyzing the ω-hydroxylation of capric acid and lauric acid

Article abstract of DOI:10.1271/bbb.60396

A cDNA encoding a cytochrome P450 (CYP76B9) was isolated from Petunia hybrida. Northern blot analysis revealed preferential expression of the gene in flowers and leaves. The recombinant yeast microsomes expressing CYP76B9 was allowed to react with capric acid and lauric acid as substrates. One major metabolite was produced from each fatty acid after incubation with yeast microsomes expressing CYP76B9. The metabolites were identified by gas chromatography-mass spectrometry (GC-MS) as ω-hydroxy capric acid and ω-hydroxy lauric acid. The kinetic parameters of the reactions were K _m = 9:4 μM and V_max = 13:6 mol min^-1 per mol of P450 for capric acid, and K_m = 5:7 μM and V_max = 19:1 mol min^-1 per mol of P450 for lauric acid. We found that the ω-hydroxy metabolites of capric acid and lauric acid can affect the plant growth of Arabidopsis thaliana. Plants grown in the presence of ω-hydroxy fatty acids exhibited shorter root length than control plants with the corresponding non-hydroxylated fatty acids.

Full text of DOI:10.1271/bbb.60396

Biosci. Biotechnol. Biochem., 71 (1), 104–113, 2007
Molecular Cloning of CYP76B9, a Cytochrome P450 from Petunia hybrida,
Catalyzing the !-Hydroxylation of Capric Acid and Lauric Acid
y
Hiromasa IMAISHI and Mariana PETKOVA-ANDONOVA
Research Center for Environmental Genomics, Kobe University, Rokkodaicho 1-1,
Nada-ku, Kobe 657-8501, Japan
Received July 12, 2006; Accepted September 16, 2006; Online Publication, January 7, 2007
[doi:10.1271/bbb.60396]
A cDNA encoding a cytochrome P450 (CYP76B9) was
CYP78A5, reported possibly to have a role in control-
6
)
isolated from Petunia hybrida. Northern blot analysis
revealed preferential expression of the gene in flowers
and leaves. The recombinant yeast microsomes express-
ing CYP76B9 was allowed to react with capric acid and
lauric acid as substrates. One major metabolite was
produced from each fatty acid after incubation with
yeast microsomes expressing CYP76B9. The metabolites
were identified by gas chromatography–mass spectrom-
etry (GC–MS) as !-hydroxy capric acid and !-hydroxy
lauric acid. The kinetic parameters of the reactions were
ling growth of the shoot apical meristem. Fatty acids
are major components of cell membranes and are
responsible for the majority of permeability functions
7
,8)
of membranes.
derivatives, several of which have important regulatory
Plants synthesize many fatty acid
9
)
roles such as jasmonates. The first enzyme in the
biosynthesis of jasmonic acid is allene oxide synthase,
an enzyme that converts fatty acid hydroperoxides to
allene oxides. Laudert et al. reported the cDNA cloning
and functional characterization of A. thaliana allene
ꢀ1
10)
Km ¼ 9:4 ꢀM and Vmax ¼ 13:6 mol min per mol of
P450 for capric acid, and Km ¼ 5:7 ꢀM and Vmax ¼ 19:1
oxide synthase (CYP 74A1). Mammalian fatty acid
!-hydroxylases have been extensively studied. They
ꢀ1
mol min per mol of P450 for lauric acid. We found
that the !-hydroxy metabolites of capric acid and lauric
acid can affect the plant growth of Arabidopsis thaliana.
Plants grown in the presence of !-hydroxy fatty acids
exhibited shorter root length than control plants with
the corresponding non-hydroxylated fatty acids.
belong to the CYP4 family of cytochrome P450
1
1)
species. These species are induced by compounds
2)
1
that are known to stimulate peroxisomal proliferation.
In yeast species, the CYP52 family is induced by
1
3)
aliphatic hydrocarbons.
catalyze terminal hydroxylation of n-alkanes in the
assimilation pathway.
CYP52A3 and CYP52A4
1
4)
Key words: cytochrome P450; Petunia hybrida; !-hy-
droxylation; lauric acid; capric acid
A limited number of fatty acid !-hydroxylases have
been characterized in higher plants. CYP86A1 was the
first fatty acid !-hydroxylase isolated from A. thali-
1
5)
Cytochrome P450 monooxygenases catalyze the oxi-
dative reactions of a wide variety of endogenous and
exogenous lipophilic compounds. In higher plants, P450
enzymes are mainly involved in the biosynthesis of sec-
ondary metabolites, such as fatty acids, phenylpropa-
ana. It was found to catalyze the !-hydroxylation of
saturated and unsaturated fatty acids with chain lengths
from C12 to C18. Microsomes from apical buds of pea
1
6)
seedlings hydroxylate lauric acid at the !-position.
CYP94A1 from Vicia sativa (V. sativa) is capable of
hydroxylating the methyl terminus of 9,10-epoxystearic
and 9,10-dihydroxystearic acids to form the correspond-
1
,2)
noids, terpenoids, and plant signal molecules. Normal
development in a model plant such as Arabidopsis
thaliana (A. thaliana) has three overlapping develop-
mental phases: vegetative development, inflorescence
development, and flower development.3 A few cyto-
chrome P450-mediated reactions give products neces-
sary for control of cell expansion in plants. The muta-
tions of P450 species in A. thaliana known as dwarf 4
1
7,18)
ing 18-hydroxy derivatives.
iva catalyzes the hydroxylation of myristic, lauric, and
CYP94A2 from V. sat-
)
19)
palmitic acids. The metabolites formed from lauric
and myristic acids have been identified by GC–MS as
the corresponding !- and (!-1)-hydroxy fatty acids,
and from palmitic acid as the corresponding (!-1)- and
4
,5)
20)
and lcr give dwarfed plants.
Another example is
(!-2)-hydroxy fatty acids. CYP94A5 from tobacco
The nucleotide sequence reported here is available in the DDBJ, EMBL, and GenBank Nucleotide Sequences Database under accession
no. AB265193 (CYP76B9).
y
To whom correspondence should be addressed. Tel/Fax: +81-78-803-5940; E-mail: himaish@kobe-u.ac.jp
Abbreviations: AHD, alcohol dehydrogenase; DEPC, diethyl pyrocarbonate; DTT, dithiothreitol; GC–MS, gas chromatography–mass
spectrometry; GSP, gene-specific primer; RACE-PCR, rapid amplification of cDNA-end polymerase chain reaction; RT-PCR, reverse
transcriptase-polymerase chain reaction; SSC, saline-sodium citrate; TLC, thin-layer chromatograph

CYP76B9 Catalyzes the !-Hydroxylation of Fatty Acid
105
0
0
catalyze the !-hydroxylation of 9,10-epoxystearic
0)
AIGG-3 , and primer 5B, 5 -TTI(A,G)(C,G)IGAI(T,C)-
0
acid.2 CYP78A1 from Zea mays has been found to
TI(A,C)TIC(G,T)I(A,C)A(T,C)-3 . The temperature pro-
ꢀ ꢀ
gram was 3 min at 94 C, 50 cycles of 50 s at 94 C, 50 s
2
1)
catalyze 12-monooxygenation of lauric acid.
2)
CYP703A12 and CYP92B1
23)
have been cloned
at 45 C, and 45 s at 72 C. PCR products were separated
on 2% agarose gels, and a major band was extracted on
a GeneElute agarose spin column (Supelco, Bellefonte,
PA). Purified cDNA fragments were cloned into the
EcoRI/HindIII site of pBluescript II SK+ (Stratagene,
La Jolla, CA), transformed into Escherichia coli strain
JM109, and sequenced. Several randomly selected
clones were sequenced, and the resulting sequences
were compared with those of reported plant P450
species using the Protein-protein BLAST search pro-
gram.
ꢀ
ꢀ
from P. hybrida. CYP703A1 was found to metabolize
lauric acid. CYP92B1 was also found to metabolize
lauric, linoleic, and linolenic acid. Functional analysis
of a cytochrome P450 monooxygenase, CYP86A8 in
A. thaliana, which catalyzes !-hydroxylation of fatty
acids, revealed several different roles of fatty acid !-
hydroxylation in the development of Arabidopsis thali-
,24)
ana.5
These reports show that the P450 species
involved in the in-chain hydroxylation and/or !-
hydroxylation of fatty acids participate in the growth
and development of A. thaliana. However, the cDNA
sequences and enzyme function of these P450 forms are
0
0
5 ,3 RACE-PCR method. To obtain cDNA from low-
27)
abundance mRNA, we used the RACE-PCR method,
1
,25)
poorly understood in higher plants.
To analyze the cDNA structure and enzyme function
of P450 species involved in the metabolism of fatty
acids, we attempted to isolate a novel P450 from Petunia
hybrida. In this paper, we describe the cloning of a novel
P450 species, CYP76B9, isolated from P. hybrida,
characterize its enzyme properties, and investigate the
effects of medium-chain fatty acids on plant growth,
using wild-type A. thaliana as a model higher plant.
employing the Marathon cDNA amplification kit (Clon-
þ
tech, Palo Alto, CA). Poly(A) RNA was isolated from
petunia leaves. An adopter ligated cDNA library was
made following the instruction manual. Using sequences
0
0
of the 5 -terminial region and the 3 -terminal region of
clone-8, we designed GSP-1 and GSP-2. The following
0
RACE-PCR primers were used (see Fig. 1): GSP-1, 5 -
0
CGACCTCGGATATCTATTTCTGACTCCCA-3 , and
0
GSP-2, 5 -CTAGTGAGCGTATGGGCAATTGGACG-
0
Materials and Methods
CAA-3 . PCR amplification of full-length cDNA was
performed using the adapter ligated cDNA library as a
template and one pair of GSP-1, GSP-2 primers. The
14
Materials. Radiolabeled [1- C] capric acid and [1-
C] lauric acid were obtained from Amersham Interna-
1
4
ꢀ
PCR conditions for RACE-PCR were 1 min at 94 C, 25
ꢀ
ꢀ
tional. (Buckinghamshire, UK), and !-hydroxy capric
acid and !-hydroxy lauric acid were from by Nacalai
Tesque (Kyoto, Japan). Thin-layer plates (Silica Gel
G60 F254; 0.25 mm) were from Merck (Darmstadt,
Germany). All other chemicals were purchased from
Nacalai Tesque (Kyoto, Japan).
cycles of 30 s at 94 C, 4 min at 68 C. The RACE-PCR
specific products were separated on 1% agarose gels,
specific bands were extracted on a GeneElute agarose
spin column, and the purified fragments were subcloned
into pT7Blue vector (Novagene, Madison, WI) and
sequenced.
Growth of plants. P. hybrida was grown as previously
described.22,23) A. thaliana ecotype columbia (Col-0)
was used. To analyze the physiological effects of fatty
acids on the growth of A. thaliana, plants were grown in
the presence or absence of fatty acids (capric acid, !-
hydroxy capric acid, lauric acid, and !-hydroxy lauric
acid) for 8 weeks under a 10-h light, 14-h dark
photoperiod at 25 C in a growth chamber. The pH of
culture medium was adjusted to 5.8–6.0 before auto-
claving.
Inverse PCR cloning. Genomic DNA from leaves of
22)
petunia was prepared as described by Imaishi et al.
Restriction digestion was carried out using 1 mg of
petunia genomic DNA treated with 10 units of XbaI for
ꢀ
16 h at 37 C. The restriction enzyme reaction mixture
was extracted once with an equal volume of phenol:
chloroform (1:1). DNA fragments were precipitated by
the addition of 1/10 volume of 3 M sodium acetate and
ꢀ
ꢀ
2.5 volumes of 100% ethanol and kept at ꢁ80 C for 2 h.
The DNA pellet was washed once in 70% ethanol, dried,
and resuspended in 10 ml of sterile water. For circular-
ization, 0.5 mg of the restriction fragments was diluted
to a concentration of 1 ng/ml in ligation buffer (50 mM
RT-PCR cloning of the partial sequence of the P450
species. The P450 gene was cloned by the RT-PCR
technique.2
2,23)
A Quick Micro mRNA purification kit
GE Healthcare, Milwaukee, WI) was used to isolate
Tris HC1, pH 7.4, 10 mM MgCl , 10 mM dithiothreitol,
2
(
1 mM adenosine triphosphate). The ligation reaction
was initiated by the addition of T4 DNA ligase to a
concentration of 20 unit/ml, and the reaction was
þ
poly(A) RNA from a mixture of petunia flowers,
leaves, stems, and roots. A pair of degenerate primers
ꢀ
(
Holton and Lester for random cloning of petunia P450
primers 4 and 5B) corresponding to the primers used by
2
allowed to proceed for 12 h at 15 C. The PCR reaction
6)
was performed with a mixture (50 ml) of 0.5 ng of
circularized DNA obtained as described above, 50 pmol
of each primer, 500 mM dNTPs, and 2.5 units Taq-
0
species were used: primer 4, 5 -AAGAATTCCCIGG-
(
A,G)CAIATIA(G,T)(C,T)(C,T)TICCIGCICC(A,G)A-

106
H. IMAISHI and M. PETKOVA-ANDONOVA
Primer-U
Primer-3
Primer-1
Primer-2
Primer-4
Domain A
GSP-2
GSP-1
Domain C
Domain B
Domain D
Primer-L
Fig. 1. Nucleotide Sequence and Deduced Amino Acid Sequence of CYP76B9.
Domains A, B, C, and D, which are highly conserved among P450 genes, are underlined. Arrows indicate the positions of GSP-1, GSP-2,
primer-1, primer-2, primer-3, primer-4, primer-U and primer-L. Clone-8 is designated in bold face.
polymerase (Parkin-Elmer, Norwalk, CT). We carried
out two PCR reactions with the PCR primers corre-
sponding to the N-terminal regions of IMT-8 cDNA. In
the first reaction, primers-1 and -2 were used. Then
second reaction was performed using primers-3, -4 and
products from the second PCR reaction were separated
on a 0.7% agarose gel, and DNA bands were purified
and ligated to the pT7Blue T-Vector and sequenced.
RT-PCR cloning of full-length cDNA. To obtain full-
length cDNA from petunia leaves, RT-PCR was carried
out. Poly(A) + RNA were isolated from petunia leaves
using a Quick-Prep Micro mRNA Purification Kit
(Pharmacia Biotech, Piscataway, NJ, USA). First-strand
cDNA was synthesized from 100 ng of poly(A) + RNA
using a First-Strand cDNA Synthesis Kit (Pharmacia
Biotech) according to the protocol of the manufacturer,
and 10 ng were used for PCR reactions. RT-PCR was
performed with 10 mM Tris–HCl (pH 8.3), 50 mM KCl,
2.5 mM MgCl2, 0.25 mM dNTP each, 25 pmol of
1
ml of sample from the first PCR reaction. The
nucleotide sequences of primers used are as follows:
0
(
GATGATGAAATG-3 ; primer-2, 5 -TCTTGCAAAA-
see Fig. 1): primer-1, 5 -ACTTCTCTTGCCACGACT-
0
0
0
0
ACAAGACTTAACATTTTCC-3 ; primer-3, 5 -TGAC-
0
CACCGTGATTAGTTGGCCTAATTTGAG-3 ; primer-
0 0
4
, 5 -TCCAATAGGTTTGTCCCGGACGTAGTCC-3 .
The positions of the inverse PCR primers are indicated
by arrows in Fig. 1. The cycling conditions were: 1 min
ꢀ ꢀ ꢀ
9
5 C, 2 min 65 C, 3 min 72 C, for 35 cycles. The DNA

CYP76B9 Catalyzes the !-Hydroxylation of Fatty Acid
107
primers, 2.5 units Taq-polymerase (Parkin-Elmer, Nor-
ꢀ
and the terminator. The full-length cDNA fragment of
CYP76B9 was ligated into pAUR123 after digestion
with SalI/XbaI to construct the expression plasmid
pAUR76J1. The resulting plasmids, pAUR123 and
pAUR76J1, were introduced into S. cerevisiae by the
walk, CT). The temperature program was 30 s 94 C,
ꢀ
ꢀ
2
PCR primers were used (see Fig. 1): primer-U, 5 -AT-
min 50 C, 3 min 72 C, 30 cycles. The following RT-
0
0
GGACTACGTGAATATTTTGCTGGGACTG-3 ; pri-
0
28)
mer-L, 5 -CTACAGGGGAGTTGGGATAGCTAGCA-
0
lithium chloride method of Oeda. The yeast culture
ꢀ
GAG-3 . The PCR products obtained were separated on
was grown in 2 ml of YPD medium at 30 C for 16 h,
1
% agarose gel, and the PCR product around 1.7 kbp
was extracted using a GeneElute Agarose Spin Columns
Supelco, Bellefonte, PA). Purified cDNA fragments
and the cells were collected and resuspended in 0.1 M
lithium chloride. The cells were then incubated with
1 mg of the plasmid DNA, 10 mg of denatured salmon
sperm DNA as a carrier, and 40% (w/v) polyethylene
glycol at room temperature for 30 min. Finally, the
transformed cells were cultured in YPD medium at
(
were cloned to the SmaI site on pBluescript II SK+ and
sequenced.
ꢀ
Northern blot analysis. Total RNA from petunia
leaves, flowers, roots, and stems was extracted with
Sepasol RNA I Super Reagent (Takara Shuzo, Otsu,
Japan). After isolation, RNA was precipitated with 70%
ethanol, dissolved in diethyl pyrocarbonate (DEPC)-
30 C for 60 min and spread onto YPD plates containing
0.5 mg/ml of Aureobasidine A for selection.
ꢀ
Yeast cells were cultured in YPD medium at 30 C
for 48 h.29) After shaking, yeast cells were collected by
centrifugation at 7,000 g for 10 min and washed with
0.1 M potassium phosphate buffer (pH 7.4). The collect-
ed pellets were resuspended in 40 ml of zymolyase
buffer (10 mM Tris–HCl, pH 7.5, 2 M sorbitol, 0.1 mM
DTT, 0.1 mM EDTA) containing 300 mg/ml of zymo-
lyase-100T (Seikagaku Kougyo, Tokyo), and incubated
ꢀ
water, and stored at ꢁ80 C. The RNA samples were
separated by gel electrophoresis (0.8% agarose, 1%
formaldehyde) and capillary transferred onto a Hybond-
þ
N
nylon membrane (Amersham International, Buck-
inghamshire, UK). Equal loading of RNA was checked
by ethidium bromide staining of the gel. The immobi-
ꢀ
with gentle shaking at 30 C for 1 h. After centrifuga-
ꢀ
lized RNA was hybridized at 65 C for 12 h in hybrid-
ization solution (6 ꢂ SSC, 7.5 ꢂ Denhardt’s reagent,
tion, the spheroplasts, washed with zymolyase buffer,
were resuspended in 15 ml of sonification buffer (10 mM
Tris–HCl, pH 7.5, 0.65 M Sorbitol, 0.1 mM DTT, 0.1
mM EDTA, 1 mM PMSF) and placed on ice for brief
sonication (50 W for 2 min, 3 times). The lysate was
centrifuged to precipitate cell debris. The resulting
supernatant was transferred to centrifuge tubes and
32
2
00 mg/ml salmon sperm DNA), with the P-labeled
3
89-bp PCR product of clone-8 as a probe. The hy-
bridized blots were washed twice with 2 ꢂ SSC/0.1%
SDS at room temperature for 15 min. An FLA-2000
bioimage analyzer (Fuji, Tokyo) was used to detect
radioactivity on the membrane.
ꢀ
centrifuged at a setting of 10,000 g for 60 min at 4 C.
The microsomal pellet was resuspended in 0.1 M
potassium phosphate buffer (pH 7.5) containing 10%
glycerol and 4.2 mM 2-mercaptoethanol, and stored at
Southern blot analysis. Genomic DNA from leaves of
2
2)
petunia was prepared as described by Imaishi et al.
ꢀ
EcoRI, HindIII, and DraI were used to digest the
purified genomic DNA. The digested DNA was sepa-
rated by 0.7% agarose gel electrophoresis and trans-
ꢁ80 C. Reduced CO-difference spectra of microsomal
fractions prepared from the transformed yeast cells were
2
8)
measured by the method of Oeda et al. The P450
hemoprotein content in the microsomal fraction was
estimated on using an extinction coefficient of 91
þ
ferred to a Hybond-N nylon membrane with 20 ꢂ SSC
buffer. The hybridization and washing conditions were
the same as those described for RNA blot analysis.
ꢁ
1
ꢁ1 28)
mM cm . Protein assay was carried out by the
0)
3
method of Bradford using Bio-Rad Protein Assay Kit I
(Bio-Rad Laboratories, Tokyo) with BSA as a standard.
DNA sequence and computer analysis. DNA sequenc-
ing was carried out as described in Ref. 22.
Enzyme assay. Fifty picomoles of P450 protein in a
microsomal fraction prepared from recombinant yeast
strain AH22/pAUR123 or AH22/pAUR76J1 was added
to a final volume of 400 ml of buffer solution (0.1 M po-
tassium phosphate buffer, pH 7.4, 2 mM glucose-6-phos-
phate, 1 unit of glucose-6-phosphate dehydrogenase,
Expression of CYP76B9 in S. cerevisiae. Cloning of
CYP76B9 into pAUR123 (Takara Bio, Otsu, Japan) was
performed by PCR amplification using two specific
primers to introduce a SalI site upstream of the initiation
codon and an XbaI site downstream of the stop codon.
The upstream primer sequence was AAAGTCGAC-
ATGGACTACGTGAATATTTTGSTGGGACTG, and
the downstream primer was AAAAGTCTTACAGGG-
GAGTTGGGATAGCTAGCAG.
S. cerevisiae strain AH22 cells were used for ex-
pression of CYP76B9 cDNA as host. The expression
vector pAUR123 contains the multicloning site between
the alcohol dehydrogenase (ADH) I gene promoter
1
4
14
10 mM radiolabeled [1- C] capric acid or [1- C] lauric
acid). After preincubation for 2 min at 30 C, reactions
ꢀ
were initiated by the addition of 5 mM NADPH, allowed
to proceed for 1 h, and then stopped by the addition of
40 ml of HCl. The reaction mixtures were extracted twice
with 800 ml of ethyl acetate, and the extracts were dried
and then dissolved in methanol. The total radioactivity
in the extracts was measured directly in a liquid

108
H. IMAISHI and M. PETKOVA-ANDONOVA
scintillation counter (LSC 3500, Aloka, Tokyo), using
Clear-Sol (Nakalai Tesque, Kyoto, Japan) for measure-
ment of radioactivity. Samples containing a total radio-
activity of 30,000 dpm were applied to TLC plates
precoated with silica gel 60F254 (Merck, Darmstadt,
Germany). The TLC solvent system for capric acid and
lauric acid was diethyl ether:petroleum ether:formic acid
(70:30:1 by volume). Radioactivity on the TLC plate was
measured with a FLA-2000 bioimage analyzer. Kinetic
parameters (Vmax and K ) were estimated by nonlinear
m
Hence inverse PCR was carried out to obtain
information on the amino-terminal region of IMT-8. A
0
1.1-kb fragment containing the 5 -region was isolated.
0
The full-length cDNA was obtained by RT-PCR with 5 -
0
and 3 -gene-specific primers, primer-U and primer-L,
designed on sequences of RACE-PCR and inverse PCR
products (see Fig. 1). We found that the full-length
cDNA clone called P450-8 consisted of a 1,719-bp
insert with an open reading frame encoding polypeptides
of 573 amino acids (Fig. 1). All of the structural and
functional motifs common to P450 proteins, including
the N-terminal membrane anchor sequence, the proline-
rich region, and the heme-binding region, are highly
conserved in P450-8. P450-8 was officially named
CYP76B9 by D. R. Nelson (Committee for Standardized
Cytochrome P450 Nomenclature). The similarity scores
for the amino acid sequences showed that the highest
sequence similarity of CYP76B9 was 54% with
CYP76B1, a xenobiotic-inducible 7-ethoxycoumarin
regression of the collected data (n ¼ 3) using substrate
concentrations ranging from 0.3 to 100 mM.
Identification of capric acid and lauric acid metab-
olites. For GC–MS analysis, nonradiolabeled capric acid
and lauric acid were used. Lauric acid metabolism was
assayed as described above, except that in this experi-
ment the reaction mixture was prepared in a final
volume of 5 ml. Metabolites were resolved by silica
TLC, and the solvent system was diethyl ether:petro-
leum ether:formic acid (70:30:1 by volume). The area
corresponding to polar metabolites generated from the
3
1)
O-deethylase from Helianthus tuberosus. The simi-
larities of CYP76B9 to CYP92B1 and CYP703A1, P450
species preferentially expressed in petunia flower buds,
were 30% and 24% respectively.
5
-ml reaction mixture was scraped and eluted twice
from silica with 4 ml of ethyl acetate. Eluted compounds
were evaporated under nitrogen and dehydrated over
P O and silica gel for 16 h. The dried residues were
Northern blot analysis
Northern blot analysis of CYP76B9 was carried out by
the use of clone-8 as a DNA probe for the flowers, roots,
stems, and leaves of petunia (Fig. 2). As shown in
Fig. 2, the level of transcripts hybridized with CYP76B9
was highest in the leaves. On the other hand, low levels
of mRNA for CYP76B9 were detected in stems and
roots. Thus, CYP76B9 was found to be expressed
mainly in flowers and leaves of petunias.
2
5
resuspended in 125 ml of a 5% hydrogen chloride:
methanol solution, heated at 70 C for 1 h, evaporated
ꢀ
under nitrogen, and dehydrated again over P2O5 and
silica gel for 16 h. The residues were resuspended in
1
00 ml of hexane and submitted to gas chromatography
and electron impact (70 eV) ionization mass spectrom-
etry (Hitachi Model M-9000 System (Hitachi, Tokyo)
on a 1% SE30 capillary column, 30 ꢂ 0:25 mm pro-
ꢀ
ꢀ
grammed from 130 to 250 C at 300 C/min.
Southern blot analysis
Southern blot analysis of CYP76B9 revealed three
bands after EcoRI digestion, but only one band in
restriction digests with HindIII and DraI, indicating that
the petunia genome contains one copy of the cyp76B9
gene (Fig. 3).
Effect of medium-chain fatty acids on the growth of
plants. Wild-type A. thaliana seeds were sterilized in
70% ethanol for 1 min, and then in a mixture of 5%
NaClO and 0.05% Tween-20 for 10 min. Seeds were
grown on MS medium with or without 100 mM of fatty
acids or hydroxy fatty acids for 20 d at 23 C under
continuous light.
ꢀ
Results
Cloning of full-length cDNA of a novel P450 species
RT-PCR cloning of the partial sequence of the P450
species was carried out. One PCR fragment, designated
clone-8 (389 bp, see Fig. 1), encoded a novel P450-like
sequence. The 5 - and 3 -RACE-PCR methods were
employed to obtain the partial cDNA fragments of P450
rRNA
F
L
S
R
0
0
Fig. 2. Northern Blot Analysis of CYP76B9 in Petunia.
Total RNA fractions were isolated from petunia flower buds,
leaves, stems, and roots. Ten micrograms of total RNA was applied
to each lane, and the RNA samples were separated by electro-
0
0
in petunia. A 5 -terminal fragment (1,450 bp) and a 3 -
terminal fragment (763 bp) were obtained. These prod-
ucts were subcloned for DNA sequencing. Analysis of
the deduced amino acid sequences of nested RACE PCR
þ
phoresis and transferred to a Hybond-N membrane. Lane F, total
RNA from flower bud stages 1–5; lane L, total RNA from leaves;
lane S, total RNA from stems; lane R, total RNA from roots. The
products showed that the cDNA clone, designated IMT-
0
389-bp PCR product of CYP76B9 cDNA (clone-8) was used as a
probe.
8
, lacked the 5 -region.

CYP76B9 Catalyzes the !-Hydroxylation of Fatty Acid
.050
109
0
E
H D
(
Bases)
-
-
9 416
6 557
-
-
2 322
2 027
-
564
0.000
400
450
500
Wavelength, nm
Fig. 3. Southern Blot Analysis of CYP76B9 in Petunia.
Genomic DNA prepared from petunia leaves was digested with
EcoRI (lane E), HindIII (lane H), and DraI (lane D). The probe was
the same as that for Northern blot hybridization. The full length of
CYP76B9 cDNA was used as a probe.
Fig. 4. Reduced CO-Difference Spectrum of CYP76B9 Expressed in
Yeast Cells.
Solid line, reduced CO-difference spectrum of yeast cells trans-
formed with AH22/pAUR76J1; dashed line, reduced CO-difference
spectrum of yeast cells transformed with AH22/pAUR123. The
microsomal fractions (43 mg of protein) were reduced with solid
Na2S3O4 and then treated with bubbling CO for 1 min.
Expression of CYP76B9 in a yeast
To characterize the enzymatic properties of
CYP76B9, this P450 species was expressed in the yeast
S. cerevisiae under the control of the AHD promoter.
S. cerevisiae AH22 cells were transformed with expres-
sion plasmids pAUR123 and pAUR76J1. The reduced
CO-difference spectrum of the microsomal fractions
from AH22/pAUR76J1 showed an absorption peak at
metabolite was detected after the recombinant yeast
microsomes expressing CYP76B9 were incubated with
capric acid (Fig. 5A). The metabolite was purified by
TLC, analyzed by gas chromatography, and identified as
!-hydroxy capric acid by comparison with the authentic
compound. The mass spectrum showed characteristic
4
4
49 nm (Fig. 4). The P450 content was estimated to be
3 pmol of P450 equivalent per milligram of microsomal
þ
þ
protein in each fraction. In contrast, the yeast cells
transformed with AH22/pAUR123 showed no charac-
teristic P450 spectrum. These results suggest that
CYP76B9 was functionally expressed in the microsomes
of recombinant yeast cells and was localized in the yeast
microsomes.
ions at m=z 203 (M , parent peak), 185 (M ꢁ H O),
2
þ
þ
172 (M ꢁ OCH ), 144 (M ꢁ H CO ꢁ C=O), and 74
3
3
and 87 (methyl esters of McLafferty fragment ions)
(Fig. 5B). In a similar way, the lauric acid metabolite
purified by TLC (Fig. 5C) and analyzed by GC–MS was
identified as !-hydroxy lauric acid by comparison with
the authentic compound. The mass spectrum showed
þ
Metabolism of fatty acids in yeast microsomes
expressing CYP76B9
characteristic ions at m=z 231 (M , parent peak), 213
þ
þ
(M ꢁ H O), 200 (M ꢁ OCH ), and 74 and 87 (methyl
2
3
The microsomal fractions prepared from the recombi-
nant yeast cells were used for metabolism assays of
endogenous fatty acids. The substrates and NADPH
were incubated with microsomal fractions from re-
combinant yeast strains AH22/pAUR123 and AH22/
pAUR76J1. TLC analysis of the reaction mixtures
revealed that CYP76B9 recombinant yeast microsomes
metabolized capric acid and lauric acid (Fig. 5A, C).
esters of McLafferty fragment ions) (Fig. 5D).
The enzyme properties of CYP76B9 were determined
using various substrate concentrations (from 0.3 to 10
mM). The kinetic parameters of the reactions were
ꢁ
1
Km ¼ 9:4 mM and Vmax ¼ 13:6 mol min
per mol of
P450 for capric acid, and Km ¼ 5:7 mM and Vmax ¼
ꢁ
1
19:1 mol min per mol of P450 for lauric acid (Fig. 6).
The R value (0.34) of the lauric acid metabolite was
Effects of fatty acids on plant growth and develop-
ment
f
coincident with that of !-hydroxylated lauric acid.23) No
activities were found in the microsomal fraction of
AH22/pAUR123. These results suggest that CYP76B9
expressed in recombinant yeast microsomes catalyzed
the metabolism of the C10 and C12 families of fatty
acids. GC–MS analysis led to characterization of the
metabolites of capric acid and lauric acid. A single polar
Figure 7 shows the result for A. thaliana roots treated
with !-hydroxy capric acid and !-hydroxy lauric acid.
In treatment with 1, 10, and 100 mM !-hydroxy capric
acid, the root lengths were 5.10, 3.30, and 0.41 cm,
respectively (Fig. 7), indicating that the root length was
reduced by 13.4-fold with respect to that of the control

110
H. IMAISHI and M. PETKOVA-ANDONOVA
A
B
172
144 185
S
H CO-C -CH -(CH ) -CH OH
3
2
2 7
2
129
O
MCA
origin
1
2
3
C
D
200
213
H CO-C-CH -(CH ) -CH OH
3
2
2 9
2
S
O
157
MLA
origin
1
2
3
50
100
150
200
250
Fig. 5. CYP76B9-Catalyzed Metabolism of [1-14C] Capric Acid and [1-14C] Lauric Acid.
A, TLC radiograms of capric acid metabolism in microsome fractions of AH22/pAUR123 and AH22/pAUR76J1. Lane 1, control (without
microsomes); lane 2, metabolism of capric acid with pAUR123 expressed in yeast cells; lane 3, metabolism of capric acid with pAUR76J1
expressed in yeast cells. B, GC–MS analysis of the capric acid metabolite. The metabolite of nonradiolabeled capric acid was purified by TLC,
an aliquot of the eluted product was derivatized to the methyl ester, and the methyl ester was subjected to gas chromatography. The methyl ester
was identified as !-hydroxy capric acid. C, TLC radiograms of lauric acid metabolism in microsome fractions of AH22/pAUR123 and AH22/
pAUR76J1. Lane 1, control (without microsomes); lane 2, metabolism of lauric acid with pAUR123 expressed in yeast; lane 3, metabolism of
lauric acid with pAUR76J1 expressed in yeast cells. D, GC–MS analysis of the lauric acid metabolite. The metabolite of nonradiolabeled lauric
acid was purified by TLC, an aliquot of the eluted product was derivatized to the methyl ester, and the methyl ester was subjected to gas
chromatography. The methyl ester was identified as !-hydroxy lauric acid. [1-14C] Capric acid (5 mM) and [1- C] lauric acid (5 mM) were
incubated with or without microsomes for 30 min. The solvent system was diethyl ether:petroleum ether:formic acid (70:30:1 by volume).
Arrows indicate the positions of the metabolites (MCA, MLA) and the substrate (S). The Rf value for lauric acid is 0.34 and the Rf value for capric
acid is 0.32.
14
(
5.50 cm) on medium supplemented with 100 mM !-
Discussion
hydroxy capric acid. One hundred mM !-hydroxy lauric
acid similarly reduced the root length to 8.1% of the
control (1 mM, 5.20; 10 mM, 3.20; and 100 mM, 0.45 cm;
Fig. 7). However, as shown in Fig. 7, CA and LA,
capric acid and lauric acid, did not inhibit root
elongation.
Cytochrome P450 species from plants are involved in
the oxidation of a large number of compounds, including
fatty acids. We report here the cDNA cloning of
CYP76B9 from petunia. The amino acid sequence
similarity of CYP76B9 was 30% and 24% with
3
2)

CYP76B9 Catalyzes the !-Hydroxylation of Fatty Acid
111
1
metabolism of fatty acids makes it difficult to understand
the biological role and effects of individual metabolites
generated during the oxidative cascade of saturated and
0
.9
A
0.8
0
0
0
0
0
0
0
.7
.6
.5
.4
.3
.2
.1
0
33)
unsaturated fatty acids. Cutin covers the aerial part of
leaves and fruits and acts as a barrier against mechanical
7
)
stress and pathogen attack. Cutin is generated mainly
from hydroxylated and epoxidized derivatives of me-
dium- and long-chain fatty acids such as palmitic acid,
–
0.3 –0.2 –0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
8
)
oleic acid, linoleic acid, and linolenic acid. In this
study, it was found that CYP76B9 metabolized medium-
chain fatty acids (capric acid and lauric acid). These
results therefore suggest that the role of CYP76B9 in
cutin biosynthesis is not of primary importance.
1
0
0
0
0
0
0
0
0
0
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
B
We studied the possibility that !-hydroxy capric acid
and hydroxy lauric acid affect plant development. For
this purpose, we used wild-type A. thaliana as a model
plant. Much of the progress in the genetic dissection of
plant lipid metabolism has come from extensive studies
of A. thaliana by Sommerville et al., who attempted to
understand the relationships between membrane fatty
–
0.3 –0.2 –0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1
/[S] µM-1
Fig. 6. Lineweaver–Burk Plot of !-Hydroxylation of Capric Acid
and Lauric Acid by CYP76B9.
A, !-hydroxylation of capric acid. B, !-hydroxylation of lauric
acid.
3
4)
acid composition and cell physiology. We found that
growth of A. thaliana, especially root growth and
elongation, was inhibited by the addition of !-hydroxy
capric acid and !-hydroxy lauric acid (Fig. 7). In
contrast, capric acid and lauric acid did not affect
the growth of A. thaliana roots. We speculate that
CYP76B9-catalyzed oxygenation of capric acid and
lauric acid produces derivatives that participate in
pathways that regulate plant growth phenomena. Lee
has shown that aliphatic fatty acids (oleic acid) can
cause the loss, through changes in root permeability, of
CYP92B1 and CYP703A1 respectively in petunia.
Amino acid sequence comparison showed that the
highest sequence similarity of CYP76B9 was 54% with
CYP76B1. CYP76B1 was expressed in yeast cells.
Microsomes from the transformed yeast catalyzed the O-
dealkylation of 7-ethoxycoumarin. CYP76B9 did not
metabolize 7-ethoxycoumarin (data not shown). In order
to analyze the enzyme function of CYP76B9, CYP76B9
was expressed in yeast AH22 cells. CYP76B9 expressed
in recombinant yeast cells catalyzed !-hydroxylation of
capric acid and lauric acid. Imaishi et al. previously
demonstrated the oxidation of lauric acid to !-hydroxy
3
1)
þ
þ
ꢁ
ions such as K , Ca , and NO , resulting in inhibition
3
5)
of root growth of Hordeum vulgare. We think that
there is a possibility that the difference in permeability
among lauric acid, capric acid, and these !-hydroxides
influences the root growth of A. thaliana.
2
1)
lauric acid by CYP78A1. CYP92B1 from petunia,
expressed in a yeast, converted linoleic acid to two polar
metabolites.2
We have shown that CYP76B9 is a capric acid !-
hydroxylase and a lauric acid !-hydroxylase, but the
biological roles of fatty acid monooxygenases, including
2)
The interplay of multiple enzymes in the oxidative
6
5
4
3
2
1
0
0
1
10 100
1
10 100
1
10 100
1
10 100 (µM)
C
CA
ω-CA
LA
ω-LA
Fig. 7. Effects of Fatty Acids on the Growth of Wild-Type A. thaliana.
Plants were grown for 20 d on MS medium in the presence of three concentrations (1, 10, and 100 mM) of capric acid, !-hydroxy capric acid,
lauric acid, and !-hydroxy lauric acid. C, Control, Wild-type A. thaliana grown in the absence of fatty acids. CA, Wild-type A. thaliana grown
in the presence of capric acid. !-CA, Wild-type A. thaliana grown in the presence of !-hydroxy capric acid. LA, Wild-type A. thaliana grown in
the presence of lauric acid. !-LA, Wild-type A. thaliana grown in the presence of !-hydroxy lauric acid.

112
H. IMAISHI and M. PETKOVA-ANDONOVA
this newly reported plant P450 cytochrome, are not well
understood. Additional studies on this P450 cytochrome
are needed to help clarify the biological roles of fatty
acid hydroxylases in the regulation of growth and
development in higher plants.
M. H., Role of the peroxisome proliferator-activated
receptor in cytochrome P450 4A gene regulation.
FASEB J., 10, 1241–1248 (1996).
1
3) Ohkuma, M., Zimmer, T., Toshiya, I., Schunck, W. H.,
Ohta, A., and Takagi, M., Isozyme function of n-alkane-
inducible cytochrome P450 in Candida maltosa revealed
by sequential gene disruption. J. Biol. Chem., 273, 3948–
Acknowledgments
3
953 (1998).
1
4) Scheller, U., Zimmer, T., Kargel, E., and Schunck, W.
H., Characterization of the n-alkane and fatty acid
hydroxylating cytochrome P450 forms 52A3 and 53A4.
Arch. Biochem. Biophys., 328, 245–254 (1996).
We thank Dr. David R. Nelson of Committee for
Standardized Cytochrome P450 Nomenclature for assis-
tance in assigning the CYP76B91 sequence to the
cytochrome P450 gene family. We also thank Dr. Hideo
Ohkawa of Fukuyama University and Dr. Kiyoharu
Oono of Kobe University for their comments on this
study. This work was supported by Scientific Research
Grant no. 11760066 to H.I. from the Ministry of
Education of Japan.
15) Benveniste, I., Tijet, N., Adas, F., Philipps, G., Sala u¨ n, J.
P., and Durst, F., CYP86A1 from Arabidopsis thaliana
encodes a cytochrome P450-dependent fatty acid !-
hydroxylase. Biochem. Biophys. Res. Commun., 243,
6
88–693 (1998).
1
6) Benveniste, I., Sala u¨ n, J.-P., Simon, A., Reichhart, D.,
and Durst, F., Cytochrome P450 dependent !-hydrox-
ylation of lauric acid by microsomes from pea seedlings.
Plant Physiol., 70, 122–126 (1982).
References
1
7) Tijet, N., Helvig, C., Pinot, F., Le Bouquin, R., Lesot,
A., Durst, F., Sala u¨ n, J. P., and Benveniste, I., Functional
expression in yeast and characterization of a clofibrate-
inducible plant cytochrome P450 (CYP94A1) involved
in cutin monomers synthesis. Biochem. J., 332, 583–589
(1998).
1
)
)
Schuler, M. A., Plant cytochrome P450 monooxygenas-
es. Crit. Rev. Plant Sci., 15, 235–284 (1996).
Durst, F., Biochemistry and physiology of plant cyto-
chrome P450. In ‘‘Frontiers in Biotransformation’’ Vol. 4,
Chap. 7, eds. Ruckpaul, K., and Rein, H., Akademie
Verlag, Berlin (1991).
2
18) Pinot, F., Benveniste, I., Sala u¨ n, J.-P., Loreau, O., Noel,
J. P., Schreiber, L., and Durst, F., Production in vitro by
the cytochrome P450 CYP94A1 of major C18 cutin
monomers and potential messengers in plant-pathogen
interactions: enantioselectivity studies. Biochem. J., 342,
27–32 (2000).
19) Le Bouquin, R., Pinot, F., Benveniste, I., Sala u¨ n, J.-P.,
and Durst, F., Cloning and functional characterization of
CYP92A2, a medium chain fatty acid hydroxylase from
Vicia sativa. Biochem. Biophys. Res. Commun., 261,
156–162 (1999).
20) Le Bouquin, R., Skrabs, M., Kahn, R., Benveniste, I.,
Sala u¨ n, J.-P., Schreiber, L., Durst, F., and Pinot, F.,
CYP94A5, a new cytochrome P450 from Nicotiana
tabacum is able to catalyze the oxidation of fatty acids to
the !-alcohol and to the corresponding diacid. Eur. J.
Biochem., 268, 3083–3090 (2001).
21) Imaishi, H., Matsuo, S., Sawai, E., and Ohkawa, H.,
CYP78A1 preferentially expressed in developing inflor-
escence of Zea mays encoded a cytochrome P450-
dependent lauric acid 12-monooxygenase. Biosci. Bio-
technol. Biochem., 64, 1696–1701 (2000).
3
)
)
Poethig, R. S., Phase change and the regulation of shoot
morphogenesis in plants. Science, 250, 923–930 (1990).
Choe, S., Dilkes, B. P., Fujioka, S., Takatsuto, S.,
Sakurai, A., and Feldman, K. A., The DWF4 gene of
Arabidopsis encodes a cytochrome P450 that mediates
multiple 22alpha-hydroxylation steps in brassinosteroid
biosynthesis. Plant Cell, 10, 231–243 (1998).
4
5
)
Wellesen, K., Durst, F., Pinot, F., Benveniste, I.,
Nettesheim, K., Wisman, E., Steiner-Lange, S., Saedler,
H., and Yephremov, A., Functional analysis of the
LACERATA gene of Arabidopsis provides evidence for
different roles of fatty acid !-hydroxylation in develop-
ment. Proc. Natl. Acad. Sci. USA, 98, 9694–9699 (2001).
Zondlo, S., and Irish, V., CYP78A5 encodes a cyto-
chrome P450 that marks the shoot apical meristem
boundary in Arabidopsis. Plant J., 19, 259–268 (1999).
Kolattukudy, P. E., Biosynthesis of a hydroxy fatty acid
polymer, cutin: identification and biosynthesis of 16-
oxo-9- or 10-hydroxypalmitic acid, a novel compound in
Vicia faba. Biochemistry, 13, 1354–1363 (1974).
Soliday, C. L., and Kolattukudy, P. E., Biosynthesis of
cutin: !-Hydroxylation of fatty acids by a microsomal
preparation from germinating Vicia faba. Plant Physiol.,
6
)
)
7
8
)
)
22) Imaishi, H., Matsumoto, Y., Ishitobi, U., and Ohkawa,
H., Encoding of a cytochrome P450-dependent lauric
acid monooxygenase by CYP703A1 specifically ex-
pressed in the floral buds of Petunia hybrida. Biosci.
Biotechnol. Biochem., 63, 2082–2090 (1999).
59, 1116–1121 (1997).
9
Weber, H., Fatty acid-derived signals in plants. Trends
Plant Sci., 7, 217–227 (2002).
1
0) Laudert, D., Pfannschmidt, U., Lottspeich, F., Hollander-
Czytko, H., and Weiler, E. W., Cloning, molecular and
functional characterization of Arabidopsis thaliana al-
lene oxide synthase (CYP 74), the first enzyme of the
octadecanoid pathway to jasmonates. Plant Mol. Biol.,
23) Petkova-Andonova, M., Imaishi, H., and Ohkawa, H., A
novel cytochrome P450, CYP92B1, specifically ex-
pressed in petunia flower buds catalyzes the monoox-
idation of long-chain fatty acids. Biosci. Biotechnol.
Biochem., 66, 1819–1828 (2002).
3
1, 323–335 (1996).
24) Benveniste, I., Saito, T., Wang, Y., Kandel, S., Huang,
H., Pinot, F., Kahn, R. A., Salaun, J. P., and Shimoji, M.,
Evolutionary relationship and substrate specificity of
Arabidopsis thaliana fatty acid !-hydroxylase. Plant
1
1
1) Simpson, A., The cytochrome P450, 4 (CYP4) family.
Gen. Pharmacacol., 28, 331–359 (1997).
2) Johnson, E. F., Palmer, C. N. A., Griffin, K. J., and Hsu,

CYP76B9 Catalyzes the !-Hydroxylation of Fatty Acid
113
Sci., 170, 326–338 (2006).
30) Bradford, M., A rapid and sensitive method for the
quantification of microgram quantities of protein utiliz-
ing the principle of protein dye binding. Anal. Biochem.,
72, 248–254 (1976).
31) Batard, Y., Le Ret, M., Schalk, M., Robineau, T., Durst,
F., and Werck-Reichhart, D., Molecular cloning and
functional expression in yeast of CYP76B1, a xeno-
biotic-inducible 7-ethoxycoumarin O-deethylase from
Helianthus tuberosus. Plant J., 14, 111–120 (1998).
32) Sala u¨ n, J. P., and Helvig, C., Cytochrome P450-depend-
ent oxidation of fatty acids. Drug Metab. Drug Interact.,
12, 261–283 (1995).
2
2
5) Helvig, C., Alayrac, C., Mioskowski, C., Koop, D.,
Poullain, D., Durst, F., and Sala u¨ n, J.-P., Suicide
inactivation of cytochrome P450 by midchain and
terminal acetylenes: a mechanistic study of inactivation
of a plant lauric acid !-hydroxylase. J. Biol. Chem., 272,
414–421 (1997).
6) Holton, T., and Lester, D., Cloning of novel P450 gene
sequences via polymerase chain reaction amplification.
In ‘‘Methods in Enzymology’’ Vol. 272, eds. Johnson, E.,
and Waterman, M., Academic Press, San Diego, pp. 275–
283 (1996).
2
2
2
7) Hofman, M. A., and Brian, D. A., A PCR-enhanced
0
33) Adas, F., Salaun, J.-P., Berthou, F., Picart, P., Simon, B.,
and Amet, Y., Requirement for ! and (! ꢁ 1)-hydrox-
ylations of fatty acids by human cytochrome P450 2E1
and 4A11. J. Lipid Res., 40, 1990–1997 (1999).
34) Sommerville, C., and Browse, J., Plant lipids: metabo-
lism, mutants, and membranes. Science, 252, 80–87
(1991).
method for determining the 5 end sequence of mRNAs.
PCR Methods Appl., 1, 43–45 (1991).
8) Oeda, K., Sasaki, T., and Ohkawa, H., Expression of rat
liver cytochrome P-450MC cDNA in Saccharomyces
cerevisiae. DNA, 4, 203–210 (1985).
9) Sakaki, T., Oeda, K., Miyoshi, M., and Ohkawa, H.,
Characterization of rat liver cytochrome P450MC syn-
thesized in Saccharomyces cerevisiae. J. Biochem., 98,
35) Lee, R. B., Effects of organic acids on the loss of ions
from barley roots. J. Exp. Bot., 28, 578–587 (1977).
167–175 (1985).