Cloning of polyketide synthase genes from amphidinolide-producing dinoflagellate Amphidinium sp.
ABSTRACT Cloning of polyketide synthase (PKS) gene for amphidinolide biosynthesis was attempted from a dinoflagellate Amphidinium sp. (strain Y-42). Fourteen beta-ketoacyl synthase gene fragments were obtained by Polymerase Chain Reaction (PCR) amplification from degenerated primer sets designed on the basis of the conserved amino acid sequences of beta-ketoacyl synthase domains in known type I PKSs. The PCR analysis using primer sets designed from these fourteen beta-ketoacyl synthase gene fragments revealed that these DNA sequences exist only in the dinoflagellates producing amphidinolides. The DNA sequence of the positive clone, which was isolated from genomic DNA library of Amphidinium sp. (strain Y-42) by PCR detection using the specific primer set, was analyzed by shotgun sequencing. The deduced gene products in the positive clone showed similarity with beta-ketoacyl synthase (KS), acyl transferase (AT), dehydratase (DH), ketoreductase (KR), and acyl carrier protein (ACP) in known type I PKSs and thioesterase (TE).
- SourceAvailable from: David Yellowlees[show abstract] [hide abstract]
ABSTRACT: Genes encoding ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) were cloned from dinoflagellate symbionts (Symbiodinium spp) of the giant clam Tridacna gigas and characterized. Strikingly, Symbiodinium Rubisco is completely different from other eukaryotic (form I) Rubiscos: it is a form II enzyme that is approximately 65% identical to Rubisco from Rhodospirillum rubrum (Rubisco forms I and II are approximately 25 to 30% identical); it is nuclear encoded by a multigene family; and the predominantly expressed Rubisco is encoded as a precursor polyprotein. One clone appears to contain a predominantly expressed Rubisco locus (rbcA), as determined by RNA gel blot analysis of Symbiodinium RNA and sequencing of purified Rubisco protein. Another contains an enigmatic locus (rbcG) that exhibits an unprecedented pattern of amino acid replacement but does not appear to be a pseudogene. The expression of rbcG has not been analyzed; it was detected only in the minor of two taxa of Symbiodinium that occur together in T. gigas. This study confirms and describes a previously unrecognized branch of Rubisco's evolution: a eukaryotic form II enzyme that participates in oxygenic photosynthesis and is encoded by a diverse, nuclear multigene family.The Plant Cell 04/1996; 8(3):539-53. · 9.25 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Regulation and evolution of dinoflagellate luciferases are of particular interest since the enzyme is structurally unique and bioluminescence is under circadian control. In this study, three new members of the dinoflagellate luciferase gene family were identified and characterized from Pyrocystis lunula. These genes, lcfA, lcfB, and lcfC, also exhibit the unusual structure and organization previously reported for the luciferase gene of a related dinoflagellate, Lingulodinium polyedrum: three repeated domains, each encoding an active catalytic site, multiple gene copies, and tandem organization. The histidine residues involved in the pH regulation of L. polyedrum luciferase activity, and implicated in the regulation of flashing, are also fully conserved in P. lunula. The interspecific conservation between the individual luciferase domains of P. lunula and L. polyedrum is higher than among domains intramolecularly, indicating that this unique gene structure arose through duplication events that occurred prior to the divergence of these dinoflagellates. However, P. lunula luciferase genes differ from L. polyedrum in several respects, notably, the occurrence of an intron in one gene (lcfC), a 2.25-kb intergenic region connecting lcfA and lcfB, and, of particular interest, an invariant rate of synonymous (silent) substitutions along the repeat domains, in contrast to L. polyedrum luciferase, where the occurrence of synonymous substitutions is practically absent in the central region of the domains.Biochemistry 01/2002; 40(51):15862-8. · 3.38 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: This review covers the structures, synthesis, biosynthesis, and bioactivity of a series of cytotoxic macrolides, named amphidinolides, isolated from symbiotic marine dinoflagellates of the genus Amphidinium which were separated from inside cells of marine flatworms. The structures of long-chain polyketides such as colopsinols isolated from Amphidinium sp. are also described.Natural Product Reports 03/2004; 21(1):77-93. · 10.18 Impact Factor
Marine dinoflagellates, a diverse group of unicellular eu-
karyotes, have been recognized as a rich source of poly-
ketides with interesting biological activity and unique car-
bon skeletons.1)Amphidinolides are a group of cytotoxic
macrolides isolated from marine dinoflagellates Amphi-
dinium sp., which are symbionts of Okinawan marine acoel
flatworms Amphiscolops spp.2)Amphidinolides have unique
structural features as follows. Although naturally occurring
macrolides generally possess even-numbered macrocyclic
lactone rings, more than half of the amphidinolides have
odd-numbered lactone rings. Exo-methylene unit and vici-
nally located one-carbon branches, which were frequently
found in amphidinolides, are also unique structural features
of these macrolides. In our continuous studies of the biosyn-
thesis of amphidinolides, incorporation patterns of 13C-la-
beled acetate for amphidinolides B, C, G, H, J, T, W, X, and
Y were investigated.2)The incorporation patterns for am-
phidinolides revealed that the main chain of these macrolides
were generated from unusual units derived only from C-2 of
acetates in addition to successive polyketide chains. The ex-
periments also revealed that all C1branched carbons were de-
rived from C-2 of acetates, and attached to C-1 of intact ac-
etate or isolated C-2 of acetate (Fig. 1). These unusual incor-
poration patterns, which might be generated from non-suc-
cessive mixed polyketide biosynthesis, could be found in
most dinoflagellate polyketides of which biosyntheses have
been studied so far.3)Though incorporation patterns of 13C-
labeled carbon and 18O-labeled oxygen for dinoflagellate
polyketides have been reported, study about biosynthesis
gene and enzyme for dinoflagellate polyketide has not re-
ported so much. Recently, approximately 700-bp DNA frag-
ments homologous with KS domains in known type I PKSs
were amplified from seven different species of dinoflagellates
by PCR.4)However, no gene homologous with other domains
in known type I PKSs has been reported until now. Here we
describe the cloning of 36.4-kb DNA fragment, which in-
cludes putative amphidinolide biosynthesis genes such as b-
ketoacyl synthase (KS), acyl transferase (AT), dehydratase
(DH), ketoreductase (KR), acyl carrier protein (ACP), and
thioesterase (TE), isolated from genomic DNA of amphidi-
nolide-producing dinoflagellate Amphidinium sp. (strain Y-
MATERIALS AND METHODS
were unialgally cultured in a seawater medium enriched with
1% ES supplement containing antibiotics (penicillin G, 50
unit/ml; streptomycin sulfate, 50unit/ml) to prevent contami-
nation, were suspended in 15ml of buffer containing 0.1 M
NaCl, 1mM Na2EDTA, and 50mM Tris–HCl (pH 7.5).
Sodium dodecyl sulfate (75mg) and proteinase K (1.5mg)
were added and the suspension was mixed by gentle inver-
sion. After incubation at 50°C for 6h, the suspension was
gently mixed with equal volume of phenol–chloroform and
incubated in the ice for 30min. The mixture was centrifuged
at 2000?g for 20min at 4°C and the supernatant was trans-
ferred to new tube. DNA was precipitated by addition of 1/15
volume of 3 M sodium acetate (pH 5.2) and two volumes of
cold ethanol and then spooled onto a glass rod. The spooled
Frozen cells of dinoflagellate, which
1314 Vol. 29, No. 7
Biol. Pharm. Bull. 29(7) 1314—1318 (2006)
Cloning of Polyketide Synthase Genes from Amphidinolide-Producing
Dinoflagellate Amphidinium sp.
Takaaki KUBOTA, Yoshiro IINUMA, and Jun’ichi KOBAYASHI*
Graduate School of Pharmaceutical Sciences, Hokkaido University; Sapporo 060–0812, Japan.
Received January 16, 2006; accepted April 21, 2006
Cloning of polyketide synthase (PKS) gene for amphidinolide biosynthesis was attempted from a dinoflagel-
late Amphidinium sp. (strain Y-42). Fourteen b b-ketoacyl synthase gene fragments were obtained by Polymerase
Chain Reaction (PCR) amplification from degenerated primer sets designed on the basis of the conserved amino
acid sequences of b b-ketoacyl synthase domains in known type I PKSs. The PCR analysis using primer sets de-
signed from these fourteen b b-ketoacyl synthase gene fragments revealed that these DNA sequences exist only in
the dinoflagellates producing amphidinolides. The DNA sequence of the positive clone, which was isolated from
genomic DNA library of Amphidinium sp. (strain Y-42) by PCR detection using the specific primer set, was ana-
lyzed by shotgun sequencing. The deduced gene products in the positive clone showed similarity with b b-ketoacyl
synthase (KS), acyl transferase (AT), dehydratase (DH), ketoreductase (KR), and acyl carrier protein (ACP) in
known type I PKSs and thioesterase (TE).
polyketide synthase; amphidinolide; dinoflagellate; Amphidinium sp.
© 2006 Pharmaceutical Society of Japan
∗ To whom correspondence should be addressed.e-mail: firstname.lastname@example.org
Fig. 1. Structure and Labeling Pattern of Amphidinolide H
DNA was rinsed with 70% ethanol and dissolved in 5ml of
Tris–EDTA buffer at 4°C.
Escherichia coli DH10B and Es-
cherichia coli EPI300 (Epicentre) were used as a host for
PCR products cloning and fosmid library construction,
respectively. Escherichia coli harboring pBluescript II
SK(?) (Stratagene) and pCC1FOS (Epicentre) were grown
overnight in LB (Luria-Bertani) medium with ampicillin and
chloramphenicol at the concentration of 100mg/ml and
Plasmid preparations, DNA restric-
tion digests, ligation reaction, and gel electrophoresis were
done by standard procedure. DNA fragments were isolated
from agarose gel by QIAquik Gel extraction Kit (Qiagen).
Plasmids for sequencing reaction were purified by QIAGEN
Plasmid Mini Kit (Qiagen).
RNA Isolation and Generation of Complementary
Total RNA was isolated from freshly cultured cells
using RNeasy Mini Kit (Qiagen). Each step was done ac-
cording to the protocols provided with the kit. Isolated total
RNA was reverse-transcribed by omniscript reverse tran-
scriptase (Qiagen) using random primer (Invitrogen).
Taq DNA polymerase (Promega)
was used for PCR amplification with Taq DNA polymerase
buffer (Promega), MgCl2, deoxynucleoside triphosphate mix-
ture, DMSO, primer set, and DNA template. PCR mixtures
were preheated at 94°C for 5min and polymerase chain reac-
tion was performed (1min at 94°C for denaturation, 30s at
40—65°C for annealing, 1min at 72°C for extension) fol-
lowed by final incubation at 72°C for 5min.
Fosmid Library Construction
DNA isolated from the dinoflagellate Amphidinium sp.
(strain Y-42) was partially digested with Sau3AI, end re-
paired, size selected, and ligated into the pCC1FOS (Epicen-
tre). The ligated DNA were packaged into MaxPlax Lambda
Packaging Extracts and introduced into Escherichia coli
EPI300 (Epicentre) by infection. Each step was done accord-
ing to the protocols provided with CopyControl Fosmid Li-
brary Production Kit (Epicentre).
The fosmid library was spread on the LB agar
plates at the concentrations to obtain 500 colonies per plate.
Colonies on each plate were mixed and the fosmids were ex-
tracted. The fosmids from each pool was screened by PCR
using specific primer sets designed based on the conserved
region. The positive pool was spread on the LB agar plates at
the concentrations to obtain less number of colonies and
screened until single positive colony was identified.
High molecular weight
pY42-F1 was performed by Macrogen Inc.
Shotgun sequencing of the fosmid
Amplification, Cloning, and Sequencing of the b b-Ke-
toacyl Synthase Domain DNA from Dinoflagellate Amphi-
dinium sp. (Strain Y-42)
The degenerated primer set,
TKKSF5 (5?-ATIGAYCCICARCARMG-3?) and TKKSR5
(5?-GTICCIGTICCRTGIGYYTC-3?), was designed on the
basis of the conserved amino acid sequences of b-ketoacyl
synthase domains in known type I PKSs to amplify the b-ke-
toacyl synthase gene from the genomic DNA of amphidino-
lide producer. The 750-bp fragments amplified from genomic
DNA isolated from the dinoflagellate Amphidinium sp.
(strain Y-42) using the degenerated primer set, were cloned
into pBluescript II SK(?) and their DNA sequences were de-
termined.9)The analysis of these DNA fragments using
BLAST (Basic Local Alignment Search Tool) of the Na-
tional Center for Biotechnology Information revealed that the
deduced amino acid sequences of fourteen 750-bp fragments
(clones pY42-41, pY42-44, pY42-48, pY42-49, pY42-51,
pY42-53, pY42-54, pY42-55, pY42-127, pY42-128, pY42-
130, pY42-132, pY42-141, and pY42-143) had significant
similarity with b-ketoacyl synthase domains in known type I
PKS. The DNA sequences of these 750-bp fragments were
similar to each other.
PCR Detection of the Specific b b-Ketoacyl Synthase
Gene from Dinoflagellates
These specific b-ketoacyl syn-
thase gene fragments were used to amplify products from fif-
teen dinoflagellates. The primer set, TKKSF6 (5?-CTTCAT-
CAGCTTGAGCTGAG-3?) and TKKSR16 (5?-TTCTCG-
AAGTGACTGCAG AG-3?), was designed to detect the mid-
dle of these unique 750-bp DNA fragments. The specific
primer set was used for the PCR using genomic DNA iso-
lated from thirteen marine dinoflagellates as templates. The
DNA fragments of anticipated size (650-bp) were amplified
from five strains of Amphidinium sp. (strain Y-5, Y-42, Y-71,
Y-72, and Y-100) from each of which amphidinolides were
isolated.2,10)While Amphidinium sp. (strain Y-52), two strains
of Amphidinium carterae (strain MAE-10 and ZAM-12), a
strains of Amphidinium elegans (strain AKA-6), two strains
of Amphidinium operculatum (strain OGA-16 and SES-5),
and two strains of Symbiodinium sp. (strain Y-79 and Y-80)
did not yield the DNA fragment of the anticipated size (Fig.
2). Though the long chain polyhydroxyl polyketides named
luteophanols have been isolated from Amphidinium sp.
L: DNA ladder, Lane 1: Amphidinium sp. (strain Y-5), Lane 2: Amphidinium sp. (strain Y-42), Lane 3: None, Lane 4: Amphidinium sp. (strain Y-52), Lane 5: Amphidinium sp.
(strain Y-71), Lane 6: Amphidinium sp. (strain Y-72), Lane 7: Amphidinium sp. (strain Y-100), Lane 8: Amphidinium carterae (strain MAE-10), Lane 9: Amphidinium carterae
(strain ZAM-12), Lane 10: Amphidinium elegans (strain AKA-6), Lane 11: None, Lane 12: Amphidinium operculatum (strain OGA-16), Lane 13: Amphidinium operculatum
(strain SES-5), Lane 14: Symbiodinium sp. (strain Y-79), Lane 15: Symbiodinium sp. (strain Y-80).
PCR Products Amplified from Genomic DNAs of Dinoflagellates Using Primer Set TKKSF6/TKKSR16
(strain Y-52),11—13)no amphidinolide has isolated from these
eight strains of dinoflagellates so far. The RT-PCR using the
specific primer sets was also attempted to confirm the ex-
pression of these specific b-ketoacyl synthase genes as
mRNAs. The DNA fragments of anticipated size were ampli-
fied from the complementary DNA generated from total
RNA isolated from freshly cultured cells of dinoflagellates
Amphidinium sp. (strain Y-42) (Fig. 3). These results sug-
gested that the unique 750-bp DNA fragments might be re-
sponsible for amphidinolide biosynthesis and encouraged us
to clone the larger PKS gene from the genomic DNA library
of Amphidinium sp. (strain Y-42).
Cloning and Sequence Analysis of the PKS Genes from
Genomic DNA Library of Amphidinium sp. (Strain Y-42)
The fosmid genomic DNA library was constructed from total
DNA of the dinoflagellate Amphidinium sp. (strain Y-42).
Screening of approximately one hundred thousand colonies
using the specific primer set, TKKSF6 and TKKSR16,
yielded one positive fosmid clone, named pY42-F1. The
shotgun sequencing of the fosmid pY42-F1 revealed that the
insert DNA consisted of 36378 nucleotides and the GC % of
it was 47.4%. The BLAST analysis of six open reading
frames translated from the nucleotide sequence revealed that
some part of the pY42-F1 insert DNA showed high similarity
with b-ketoacyl synthase (KS), acyl transferase (AT), dehy-
dratase (DH), ketoreductase (KR), acyl carrier protein (ACP)
motifs in known type I PKS genes and thioesterase (TE)
genes (Fig. 4, Table 1). Two amino acid sequences (translated
from nucleotide 14587—15816 and 23740—24081) found in
5?3? frame 1 showed high similarity with KS(latter half)/
AT(front half) domains in known type I PKS and TE(front
half), respectively. Three amino acid sequences (translated
from nucleotide 15818—16972, 20933—22771, and 24521—
24928) found in 5?3? frame 2 showed high similarity with
AT(latter half)/DH domains, KR/ACP domains, and TE(lat-
ter half), respectively. One amino acid sequence (translated
from nucleotide 13911—14627) found in 5?3? frame 3
showed high similarity with KS(front half) domain. DNA se-
quence of 750-bp fragment in the pY42-141 was identical
with that of putative KS gene in pY42-F1. On the other hand,
DNA sequences of other thirteen 750-bp fragments were
slightly different from that of putative KS gene in pY42-F1
These resuts indicated that these fragments corresponded to
KS gene for other modules. Comparison of the active-site se-
quence of the AT domain with the known malonyl-CoA and
methylmalonyl-CoA acyltransferase genes indicated that the
AT domain found in the 36.4-kbp fragment might belong to
malonyl-CoA acyltransferase.14)This result suggested that
frame shifts occurred in the middle of KS, AT, and TE genes,
and the introns might exist in the middle of these genes.
Though there were approximately 4000 nucleotides between
DH and KR genes, no significant similarity with known
ORFs was found between these genes. This result suggested
that if nucleotide sequences encode KS/AT/DH and KR/ACP
were translated to one PKS, an intron might exist between
DH and KR genes.
One of the most attractive issues of amphidinolide biosyn-
thesis is the truncation of C-1 of intact acetate in main chain.
Several hypotheses that explain the truncation of the carbon
chain have been proposed.15)The participation of tricar-
1316Vol. 29, No. 7
Table 1.Deduced Function of Amino Acid Sequences in the 36.4 kbp DNA Fragment from Dinoflagellate Amphidinium sp. (Strain Y-42)
Identity similarity Accession No.
PKS (KS domain)
PKS (KS/AT domain)
PKS (AT/DH domain)
PKS (KR/ACP domain)
Type II thioesterase
StiG, Stigmatella aurantiaca
StiA, Stigmatella aurantiaca
Putative PKS, Nocardia farcinica IFM 10152
EPOS C, Polyangium cellulosum
Mcy T, Planktothrix agardhii
ORF36, Nonomuraea sp. ATCC 39727
Fig. 4. Map of the 36 kb DNA Fragment from Dinoflagellate Amphidinium sp. (Strain Y-42)
DNA of Dinoflagellates Amphidinium sp. (Strain Y-42)
L: DNA ladder, Lane 1: total RNA, Lane 2: complementary DNA.
PCR Products Amplified from Total RNA and Complementary
boxylic acid (TCA) cycle was the most likely candidate.
However, this possibility could be excluded, since the incor-
poration rate of 13C-labeled carbon at isolated C-2 of acetates
might be lower than intact acetates, if these carbons were de-
rived from acetate via TCA cycle. But the feeding experi-
ments revealed that the incorporation rates of isolated C-2 of
acetates were identical with C-2 of intact acetates. As a sec-
ond candidate, the occurrence of a Favorskii or benzil-ben-
zilic acid rearrangement in the middle of the chain was pro-
posed. This hypothesis could explain that incorporation rates
of all 13C-labeled carbon were approximately same. However,
these rearrangements are not enough to complete the trunca-
tion. Some additional decarboxylation-like enzyme might be
necessary to remove C-1 of acetate, which could remain as a
branched carbon. A new mechanism for the formation of the
truncated polyketide backbones has been reported as
follows.15)A chain extension could be performed by conden-
sation with an extender malonate unit and the resulting b-ke-
toester could be reduced and dehydrated to give an elongated
unsaturated polyketide chain. The unsaturated polyketide
chain could be then epoxidized and decarboxylated to give a
truncated aldehyde. Finally the aldehyde could be re-oxi-
dized to a carboxylic acid and transferred into the next exten-
sion cycle. The one carbon extension and truncation could be
performed by one multifunction enzyme possessing all do-
mains necessary to complete these steps. Alternatively, con-
densation, reduction, and dehydration could be performed by
type I PKS, then epoxidation, decarboxylation, and re-oxida-
tion could be performed by other enzyme(s). The deduced
products of two amino acid sequences, found in 36.4-kbp ge-
nomic DNA from the dinoflagellate Amphidinium sp. (strain
Y-42), showed similarity to known type II TE. The type II TE
genes encode mono-function enzyme, and could be found
with PKS or non-ribosomal peptide synthase (NRPS) genes.
Though one of the functions of this sort of TE was proposed
to hydrolyze misacylated aberrant intermediates that might
block the PKS or NRPS.16)However, the function of this type
II TE could be responsible for the release of unsaturated
polyketide chain from the ACP domain for the truncation
Dinoflagellates are a diverse group of unicellular eukary-
otes with large and unusual genome. The amounts of DNA
per haploid nucleus are up to 60 times larger in comparison
with human. And the chromosomes are permanently con-
densed and the chromatin structure is different from other
eukaryotes because of lacking of typical histones.17,18)In this
study, a fosmid pY42-F1, which includes 36.4-kbp genomic
DNA fragment from amphidinolide-producing dinoflagellate
Amphidinium sp. (strain Y-42), has been found from approxi-
mately 100000 clones. The fragment included six open read-
ing frames homologous with known KS, AT, DH, KR, ACP,
and TE genes. Though the total size of six open reading
frames was only 5625-bp and these amino acid sequences
exist in the middle of 36.4-kbp fragment, no other sequence
showing significant similarity with known polyketide biosyn-
thesis genes was found on the both side of them. Amphidino-
lides H, for example, is twenty-six memberd macrolide and
the main chain consists of sixteen units derived from acetate.
However, if deduced products of these genes could not work
interactively, they might be responsible for only one exten-
sion step performed by condensation with a malonate unit.
This represents the difficulty of cloning of whole amphidino-
lide biosynthesis gene from genomic DNA library. Occur-
rence of intron is obviously one of the reasons that make it
difficult to clone the amphidinolide biosynthesis gene from
genomic DNA. Though the gene of dinoflagellate does not
always contain introns,19)several examples about the occur-
rence of introns in dinoflagellate genes have been re-
ported.20—22)Six introns in three rubisco genes of Symbio-
dinium sp., four introns in DNA-binding protein HCc gene of
Crypthecodinium cohnii, and one intron in luciferase gene
lcfC of Lingulodium polyedrum have been reported. These
dinoflagellate introns do not have the conserved nucleotides
sequences found in known group I and II introns of other or-
ganisms. And these dinoflagellate introns do not share com-
mon nucleotides sequences at their putative splicing sites
neither.23)The construction of complimentay DNA library of
amphidinolide-producing dinoflagellate might be the best al-
ternative to clone amphidinolide biosynthesis genes. And the
comparison of genes from genomic DNA and complimentay
DNA would also give us further information about dinofla-
gellate intron and spacer region.
G. Floss, Yutaka Ebizuka, and Dr. Isao Fujii for helpful dis-
cussions. We are grateful to Dr. Takeo Horiguchi for provid-
ing the genomic DNA of dinoflagellates. We thank Dr.
Masashi Tsuda, Ms. Naoko Izui, and Ms. Rie Iwamoto for
technical assistance. This work was supported by Grant-in-
Aid for Science Research from the Ministry of Education,
Culture, Sports, Science, and Technology of Japan.
We wish to thank Professors Heinz
1)Rein K. S., Borrone J., Comp. Biochem. Physio., B124, 117—131
Kobayashi J., Tsuda M., Nat. Prod. Rep., 21, 77—93 (2002).
Moore B. S., Nat. Prod. Rep., 16, 653—674 (1999).
Snyder R. V., Gibbs P. D. L., Palacios A., Abiy L., Dickey R., Lopez J.
V., Rein K. S., Mar. Biotechnol., 5, 1—12 (2003).
Kobayashi J., Shimbo K., Sato M., Tsuda M., J. Org. Chem., 67,
Shimbo K., Tsuda M., Izui N., Kobayashi J., J. Org. Chem., 67,
Tsuda M., Izui N., Shimbo K., Sato M., Fukushi E., Kawabata J., Ka-
tsumata K., Horiguchi T., Kobayashi J., J. Org. Chem., 68, 5339—
Tsuda M., Izui N., Shimbo K., Sato M., Fukushi E., Kawabata J.,
Kobayashi J., J. Org. Chem., 68, 9109—9112 (2003).
Marchuk D., Drumm M., Saulino A., Collins F. S., Nucl. Acids Res.,
19, 1154 (1991).
Tsuda M., Kariya Y., Iwamoto R., Fukushi E., Kawabata J., Kobayashi
J., Mar. Drugs 3, 1—8 (2005).
Doi Y., Ishibashi M., Nakamichi H., Kosaka T., Ishikawa T.,
Kobayashi J., J. Org. Chem., 62, 3820—3823 (1997).
Kubota T., Tsuda M., Doi Y., Takahashi A., Nakamichi H., Ishibashi
M., Kawabata J., Kobayashi J., Tetrahedron, 54, 14455—14464 (1999).
Kubota T., Takahashi A., Tsuda M., Kobayashi J., Mar. Drugs, 3,
Izumikawa M., Murata M., Tachibana K., Ebizuka Y., Fujii I., Bioorg.
Med. Chem., 11, 3401—3405 (2003).
Shimizu Y., Curr. Opin. Microbiol., 6, 236—243 (2003).
Heathcote M. L., Staunton J., Leadlar P. F., Chem. Biol., 8, 207—220
Schwarzer D., Mootz H. D., Linne U., Marahiel M. A., Proc. Natl.
Acad. Sci. U.S.A., 99, 14083—14088 (2002).
Spector D. L., “Dinoflagellates,” Academic Press, Orlando, 1984.
Taylor F. J. R., “The Biology of Dinoflagellates,” Blackwell Scientific,
Li L., Hastings J. W., Plant Mol. Biol., 36, 275—284 (1998).
Rowan R., Whitney S. M., Fowler A., Yellowlees D., Plant Cell, 8,
Yoshikawa T., Uchida A., Ishida Y., Fish. Sci., 62, 204—209 (1996).
Okamoto O. K., Liu L., Robertson D. L., Hastings J. W., Biochemistry,
40, 15862—15868 (2001).
1318Vol. 29, No. 7