Insecticidal compounds against mosquito larvae from Oscillatoria agardhii strain 27

Article abstract of DOI:10.1002/(SICI)1522-7278(2000)15:2<114::AID-TOX7>3.0.CO;2-P

It was found that Oscillatoria agardhii strain 27 produced compounds toxic against mosquito larvae (Aedes albopictus), therefore, these compounds were extracted and separated for insecticide development. Structural characterization of the toxic fraction by ¹H-NMR and GC-MS showed that these compounds are not the neurotoxins or hepatotoxins conventionally produced by Oscillatoria, but it contained a mixture of unsaturated fatty acids, oleic, linoleic, and γ-linolenic acids, as well as saturated fatty acids, myristic, palmitic, and stearic acids. In a bioassay developed for these hydrophobic compounds using mosquito larvae, authentic unsaturated fatty acids were shown to be toxic, whereas saturated ones were active. The results suggested that it might be possible to use unsaturated fatty acids as environmentally safe and effective insecticides without the side effects of the chemically synthesized insecticides. (C) 2000 by John Wiley and Sons, Inc.

Full text of DOI:10.1002/(SICI)1522-7278(2000)15:2<114::AID-TOX7>3.0.CO;2-P

Insecticidal Compounds Against Mosquito Larvae
from Oscillatoria agardhii Strain 27
1
2
1
1
1
2
Ken-ichi Harada, Mette Suomalainen, Hideaki Uchida, Hiroaki Masui, Kuniyo Ohmura, Jari Kiviranta,
3
4
Marja-Leena Niku-Paavola, Takaya Ikemoto
1
Faculty of Pharmacy, Meijo University, Tempaku, Nagoya 468-8503, Japan
2
Department of Pharmacy, University of Helsinki, P.O. Box 15, FIN-00014 Helsinki, Finland
VTT Biotechnology and Food Research, Tietotie 2, P.O. Box 1501, FIN-02044 Espoo, Finland
Department of Parasitology, Teikyo University School of Medicine, Itabashi, Tokyo 173-8605, Japan
3
4
Received 21 April 1999; revised 1 June 1999; accepted 3 June 1999
ABSTRACT: It was found that Oscillatoria agardhii strain 27 produced compounds toxic against
mosquito larvae ( Aedes albopictus), therefore, these compounds were extracted and separated for
1
insecticide development. Structural characterization of the toxic fraction by H-NMR and GC-MS showed
that these compounds are not the neurotoxins or hepatotoxins conventionally produced by Oscillatoria,
but it contained a mixture of unsaturated fatty acids, oleic, linoleic, and ␥-linolenic acids, as well as
saturated fatty acids, myristic, palmitic, and stearic acids. In a bioassay developed for these hydrophobic
compounds using mosquito larvae, authentic unsaturated fatty acids were shown to be toxic, whereas
saturated ones were not active. The results suggested that it might be possible to use unsaturated fatty
acids as environmentally safe and effective insecticides without the side effects of the chemically
synthesized insecticides. ᮊ 2000 by John Wiley & Sons, Inc. Environ Toxicol 15: 114᎐119, 2000
Keywords: cyanobacteria; Oscillatoria agardhii; mosquito; unsaturated fatty acid
INTRODUCTION
Most modern insecticides are synthetic chemicals
except for a group of control agents produced by Bacil-
lus thuringiensis ŽH ¨o fte and Whiteley, 1989.. Malathion
ŽMulla et al., 1981., one of the commonly used syn-
thetic insecticides, was designed to minimize the mam-
malian toxicity, because it would be detoxicated by a
mammalian enzyme. However, it will be metabolized in
insects to an active form by another enzyme. Although
the risk of adverse effects of insecticides to human
heath has been reduced from the early stages of insec-
ticide development, acute poisonings still occur. Long-
term health effects caused by insecticide bioaccumula-
tion in food chains are another concern.
Many kinds of harmful insects have caused inconve-
nience and diseases. For example, mosquitoes can
spread malaria, filarial diseases, and dengue and yellow
fevers, which are serious infections even in our modern
society. Control of mosquitoes carrying infectious dis-
eases would be much simpler and effective if the ac-
tions are directed against mosquito larvae living in
puddles instead of the flying adult ones.
Correspondence to: Ken-ichi Harada; e-mail: kiharada@meijo-u.
ac.jp
ᮊ 2000 by John Wiley & Sons, Inc.
1
14

UNSATURATED FATTY ACIDS AS AN INSECTICIDE
115
The problems related to the acute and chronic toxic-
ity and environmental pollution by insecticides are not
yet completely solved, although modern synthetic insec-
ticides have been designed to show highly selective
toxicity. Some approaches using natural enemies and
sexual pheromones to control the harmful insects were
not satisfactory due to their relatively high cost and
lack of immediate effects ŽCoats, 1994; Copping, 1996..
and structural determination of these active com-
pounds.
MATERIALS AND METHODS
Reagents
However, when a perfect insecticide is not available, it
is still possible to apply natural products as insecticides
without damaging the environment. Natural products
are produced in the environment, therefore, they prob-
ably do not cause serious environmental damage. Com-
pared to man-made pesticides, residual compounds
originating from natural products should be easily
detoxicated by microorganisms ŽLange and Lopez,
Some organic solvents, including methanol ŽMeOH.,
ethyl acetate ŽEtOAc., acetone, acetonitrile ŽCH CN.,
3
and diethyl ether were of extra pure grade and distilled
at atmospheric pressure before use. Tetrahydrofuran
ŽTHF., dimethyl sulfoxide ŽMe SO., ethanol ŽEtOH.,
2
n-butanol Žn-BuOH., N-methyl-NЈ-nitro-N-nitroso-
guanidine ŽMNNG., lauric acid, myristic acid, palmitic
acid, stearic acid, ␥-linolenic acid, linoleic acid, oleic
acid, eicosapentaenoic acid, and arachidonic acid were
of guaranteed grade. 2,6-Di-tert-butyl-p-cresol ŽBHT:
butyl hydroxytoluene., ŽBHT: butyl hydroxytoluene.,
n-decylaldehyde, n-octylaldehyde, n-hexylaldehyde, and
sorbitan monolaurate ŽSpan 20. were of extra pure
1
996.. Therefore, we attempted to isolate some active
compounds against mosquito larvae from cyanobac-
teria.
Cyanobacteria occur worldwide in eutrophic fresh
and brackish waters and they sometimes cause mam-
malian toxicoses with a number of different toxins
ŽSivonen and Jones, 1999.. Two main types of
cyanobacterial toxins, hepatotoxins and neurotoxins,
have been identified. Hepatotoxic microcystins are pro-
duced by Microcystis, Anabaena, Nostoc, and Oscillato-
ria ŽRinehart et al., 1994.. These cyclic peptides are
hepatotoxic due to their selective uptake by the liver
and inhibitory activity against protein phosphatases 1
and 2A ŽYoshizawa et al., 1990; Honkanen et al., 1990;
MacKintosh et al., 1990.. They have also been identi-
fied as tumor promoters ŽFalconer, 1991; Nishiwaki-
Matsushima et al., 1992; Ohta et al., 1994.. Another
grade. These solvents and reagents were purchased
from Nacalai Tesque ŽKyoto, Japan.. 3-wŽ3-Cholami-
dopropyl. dimethylammoniox-1-propanesulfonate
ŽCHAPS. and polyoxyethylenesorbitan monolaurate
ŽTween 20. were of extra pure grade and obtained from
Sigma Chemicals ŽSt. Louis, MO, USA.. Deuterated
chloroform, CDCl Žisotope enrichment: 99.95%., was
3
purchased from E. Merck ŽDarmstadt, Germany..
Instrumental Analysis
1
Proton nuclear magnetic resonance Ž H-NMR. spectra
group of toxins includes the neurotoxins anatoxin-a,
anatoxin-aŽs., saxitoxin, and related compounds ŽCar-
michael, 1992.. Neurotoxins have been isolated from a
variety of species, but with unrelated structures. Gener-
ally speaking, cyanobacteria have been characterized as
a rich source of bioactive compounds ŽNamikoshi and
Rinehart, 1996., for example, antimicrobial compounds
and enzyme inhibitors. Most of the bioactive molecules
are cyclic peptides, while many other types are also
found: linear peptides, phosphonates, macrolides, etc.
ŽMoore et al., 1996..
In addition to the activities described above, we have
recently discovered that Oscillatoria agardhii strain 27
could produce certain unknown active compoundŽs.
against aquatic invertebrates, mosquito Ž Aedes aegypti.,
and brine shrimp Ž Artemia salina.. The active com-
in CDCl were recorded on a JNM A-600 spectrometer
3
ŽJEOL, Tokyo, Japan.. Gas chromatography᎐mass
spectrometry ŽGC-MS. was performed using a GC-
9
ArGCMS-QP1000 ŽShimadzu, Kyoto, Japan. under
the following conditions: column: Silicone OV-101 wGas
Chrom Q, 1.1 m=3.0 mm inner diameter ŽI.D.., Wako
Pure Chemicalx, injector: CLH-702 ŽShimadzu., column
temperature: 130᎐250ЊC Žheated 5ЊCrmin., carrier gas:
He Ž40 mlrmin., separator temperature: 250ЊC, ion
source temperature: 250ЊC, ionization mode: chemical
ionization ŽCI. Žiso-butane, 200 eV., mass range: mrz
6
0᎐350, interval: 3.0 s, analytical time: 1᎐25 min.
Mosquito
Aedes albopictus was collected in Saitama Prefecture,
Japan and mosquito’s eggs were stored in a refrigerator
until use. They were maintained in water in a beaker at
25ЊC and the hatched larvae were reared until second
instar. In this bioassay the second instar larvae were
only used.
pounds originating from the strain showed a high selec-
tive toxicity to some harmful insect larvae with no
toxicity to mice, thus indicating the potential for being
developed to insecticides. ŽKiviranta and Abdel-
Hameed, 1994.. In this paper, we describe the isolation

1
16
HARADA ET AL.
Bioassay
RESULTS
Bioassay
The bioassay was performed using the larvae of
mosquito Ae. Albopictus. About 10 larvae of Ae. al-
bopictus and a sample were prepared in a water solu-
tion Ž1 mL. containing 0.01% Žvrv. Tween 20 in a glass
disposable tube Ž75=10 mm I.D.. at room temperature
under fluorescent lamp illumination. The following
control samples were used: Ža. larvae in water only and
Žb. larvae in water containing a surfactant. The degree
In a previous experiment both mosquito Ž Aedes aegypti.
larvae and newborn brine shrimps Ž Artemia salina.
were used for the bioassay ŽKiviranta and Abdel-
Hameed, 1994., but in this study, mosquito Ž Aedes
albopictus. was only used because it is easily available
in Japan and is taxonomically close to Ae. aegypti
ŽTanaka et al., 1979.. It was very important for the
bioassay to standardize the conditions to reliably evalu-
ate the active fractions. All assays were performed
using only the second instar larvae in disposable glass
tubes to unify the susceptibility and to improve the
reproducibility.
of activity was judged by the percentage of living larvae
Žcounted by the naked eye. upon exposures for up to
2
4 h.
Extraction
Lyophilized cells Ž1.3 g. of O. agardhii strain 27 were
extracted three times Žeach 30 min. with MeOH Ž1=
The MeOH extract of the lyophilized cells was first
partitioned into EtOAc and water. After concentration
of each layer, the EtOAc fraction showed the activity,
whereas the water fraction showed no activity. This
initial experiment revealed that the active compounds
have a relatively high lipophilic character, and these
could not likely be assayed in a purely aqueous system.
Therefore, we tried to select a few organic solvents and
surfactants to increase in dissolving the desired com-
poundŽs. in water. They were selected on the bases of
2
50 mL once and 2=125 mL. in an ultrasonicator
ŽBranson B-1200, Yamato, Tokyo, Japan. while cooling
with crushed ice. The extract was centrifuged at 3000
rpm Ž900=g. for 10 min and the combined super-
natant was then concentrated in vacuo. to give 165.3
mg of a dark-green oily sample. The MeOH extract was
partitioned between water Ž250 mL. and EtOAc Ž3=
2
50 mL.. The resulting EtOAc layer was evaporated
to dryness and afforded a dark-brown oily residue
having a high solubility for organic compounds and
being easily mixed with water in arbitrary proportions.
Experiments were performed to select the solvent con-
centrations that showed no toxicity. Acetone and
Ž22.2 mg..
Separation
Me SO were found to have no toxicity at concentra-
2
tions less than 2.5% Žvrv., while EtOH was toxic even
An ODS Žoctadecylsilanized. silica gel cartridge ŽSep-
ᮋ
Ž
.
Pak , 0.5 g, Waters, MA, USA. was conditioned with
at 2.5% vrv . Liquid surfactants having different
values for the hydrophile᎐lipophile balance ŽHLB.,
ŽHelenius and Simons, 1975. were also selected for use
in the bioassay. Span 20 ŽHLB value: 9.0. indicated
toxicity at 0.01% Žvrv., while Tween 20 ŽHLB value:
MeOH Ž10 mL. followed by distilled water Ž10 mL.. An
aliquot Ž2.9 mg. of the EtOAc extract was adsorbed
into the cartridge in a small amount of 70%
MeOHrdistilled water. The cartridge was rinsed with
5 mL of 70% MeOHrdistilled water and eluted Žeach
1
6.7. showed no toxicity at the same concentration
1
step 15 mL. with 80 and 90% MeOHrdistilled water,
level. Although the powdery surfactant CHAPS was
nontoxic at 0.01% Žwrv., it was not effective in dissolv-
MeOH, CH CN, and THF. The eluates were then
3
evaporated to dryness as separate fractions.
ing the active fraction. Having a low toxicity to the
mosquito larvae Tween 20 was the most appropriate
surfactant to promote the solubility of the active frac-
tion, which was determined in a preliminary test using
two model compounds, oleic acid and linoleic acid,
as shown in Table I. Therefore, all assays were carried
out with the addition of 0.01% Tween 20 in aqueous
solution.
Methylation
MNNG Žca. 50 mg. was added to a mixture of ether
Ž1 mL. and 0.5 molrl NaOH in distilled water Ž0.5 mL.
in a small glass tube while cooling in an ice bath and
stirring using a small spinning bar. An aliquot of the
resulting yellow ether solution Žstoichiometric excess.
was individually added to each mixture of saturated
fatty acids Žeach 2 mg. and unsaturated ones Žeach 2
mg. and the ether solution of the active fraction iso-
lated from the cyanobacterium in small glass tubes
while stirring with a small spinning bar at room tem-
perature.
Separation
The MeOH extract showed the highest activity among
the three tested extracts of the lyophilized cyanobacte-
ria: MeOH, H O, and 5% acetic acid aqueous solution,
2
which were efficient for the extraction of the hepato-

UNSATURATED FATTY ACIDS AS AN INSECTICIDE
117
TABLE I. Comparison of activity of fatty acids in solvents with or without surfactant
Time Žhr : min.
Concentration
No.
Sample
Solvent
ŽmgrmL.
0:00
2:30
5:30
19:30
1
2
3
4
5
6
Control 1
Control 2
Oleic acid
Oleic acid
Linoleic acid
Linoleic acid
H O
0.01% Tween 20
0
0
1
1
1
1
ᎏ
ᎏ
ᎏ
ᎏ
ᎏ
ᎏ
ᎏ
ᎏ
ᎏ
q
ᎏ
q
ᎏ
ᎏ
q
q
ᎏ
q
ᎏ
ᎏ
q
2
H O
2
0.01% Tween 20
qq"
q
H O
2
0.01% Tween 20
qq
qqq: annihilation, qq": almost annihilation, qq: half of the larvae dead, q: some larvae dead, ": unclear activity, ᎏ: no activity.
toxic microcystins ŽHarada, 1996.. To eliminate im-
Chemical Structures of Desired Compounds
and Their Activity
purities such as green pigments in the MeOH extract,
many kinds of solvent combinations were tried for
liquid᎐liquid partition. The desired active compounds
were selectively concentrated in the EtOAc layer ac-
cording to the manner shown in Fig. 1. This partition
method was remarkably effective, because the activity
of the EtOAc layer was significantly higher than that of
another fraction, which contained most of the impuri-
ties. An aliquot of the EtOAc layer was successively
separated on an ODS silica gel cartridge ŽFig. 1.. Only
1
The H-NMR spectrum ŽFig. 2, top. of the 90% MeOH
fraction suggests that this fraction contains several
unsaturated fatty acids, because the peaks at about 5.3
and 1.2 ppm are assigned to olefin and methylene
protons, respectively, which are characteristic of unsat-
urated fatty acids with a long carbon chain. The spec-
trum of ␥-linolenic acid ŽFig. 2, bottom., as a typical
two fractions, 90% MeOH aq. and THF, among the six
fractions showed activity and they were instrumental-
ly analyzed to elucidate the structures of the active
compounds.
1
Fig. 1. Separation procedure of the active compounds
Fig. 2. Comparison of H-NMR spectra of the 90% MeOH
from cyanobacteria.
fraction and an unsaturated fatty acid.

1
18
HARADA ET AL.
example of an unsaturated fatty acid, has the corre-
sponding signals to those of the active fraction. As it
was very difficult to characterize an individual fatty
1
acid by H-NMR spectroscopy, the following technique
was carried out to identify the fatty acids. Commer-
cially available fatty acids were converted to their
methyl esters using diazomethane and analyzed by GC-
CI-MS ŽFig. 3.. The 90% MeOH fraction was also
analyzed in the same manner and found to be a mix-
ture of unsaturated Žoleic, linoleic, and ␥-linolenic
acids. and saturated fatty acids Žstearic, palmitic, and
myristic acids., in which palmitic acid was predomi-
nantly detected. As shown in Table II, the unsaturated
fatty acids have activities, whereas the saturated ones
do not. Therefore, the active compounds in the active
fraction were confirmed to be a mixture of unsaturated
fatty acids. On the other hand, the active compound in
the THF fraction was identified as BHT Žbutyl hydroxy-
toluene. by instrumental analyses and a direct compari-
son with an authentic sample. Actually, the authentic
BHT sample also showed strong activity against the
mosquito larvae. Since BHT is used as an additive
against peroxide production in certain solvents such as
THF, it is not derived from the cyanobacterium.
DISCUSSION
In a previous study, an active compound in the water
soluble fraction from the same cyanobacterium strain
was toxic to the mosquito Aedes aegypti and shrimp
Artemia salina ŽKiviranta et al., 1991; Kiviranta and
Abdel-Hameed, 1994.. In contrast with the previous
results, we found that such activity is present in the
nonpolar fraction in the present study. The nonpolar
fraction was found to contain a mixture of unsaturated
fatty acids including oleic, linoleic, and ␥-linolenic acids,
which are active against the mosquito larvae. The dif-
ference between the previous and present experimental
results may be due to the extraction method used.
Many studies have been conducted on the biological
effect of lipids including fatty acids produced by
cyanobacteria. Monoglycosyl-diglyceride and diglycosyl-
diglyceride from Phormidium tenue have autolytic
properties ŽMurakami et al., 1991.. Unsaturated fatty
acids extracted from Aphanizomenon flos-aquae and
Microcystis aeruginosa inhibit the growth of the green
algae Chlorella ŽIkawa et al., 1996.. Furthermore, fatty
Fig. 3. GC-CI-MS chromatograms of methyl esters of au-
thentic fatty acids and that of the active fractions. TIC: total
ion chromatogram, MC: mass chromatogram, m/z value of
[
M+H]+ of each methyl ester in MC: saturated fatty acids,
acids from Microcystis aeruginosa had shown the poten-
q
q
m/z 215 (lauric acid); m/z 243 (myristic acid); m/z 271
palmitic acid); m/z 299 (stearic acid), unsaturated fatty
tial to inhibit potently fish gill Na rK -ATPase activ-
(
ity ŽBury et al., 1998.. Morohashi et al. Ž1991. identified
acids, m/z 293 (␥-linoleic acid); m/z 295 (linoleic acid);
m/z 317 (eicosapentenoic acid); m/z 319 (arachidonic
acid).
long-chain fatty acids as an inhibitor in brine shrimp,
Artemia salina. In most cases biological activities are
mainly due to unsaturated long-chain fatty acids rather
than to the methylated or saturated forms ŽBury et al.,

UNSATURATED FATTY ACIDS AS AN INSECTICIDE
119
TABLE II. Comparison of activity of unsaturated fatty
acids with saturated fatty acids
Carmichael, W. W. J Appl Bacteriol 1992, 72, 445᎐459.
Coats, J. R. Annu Rev Entomol 1994, 39, 489᎐515.
Copping, L. G. Crop Protection Agents from Nature: Natural
Products and Analogues; Royal Society of Chemistry: Cam-
bridge, UK, 1996; Introduction.
Concentration
ŽmgrmL.
Time
24:00
No.
Sample
1
2
3
4
5
Palmitic acid Ž16 : 0.
Stearic acid Ž18 : 0.
Oleic acid Ž18 : 1.
Linoleic acid Ž18 : 2.
␥-Linolenic acid Ž18 : 3.
1
1
1
1
1
"
ᎏ
qq
qq
qq"
Falconer, I. R. Environ Toxicol Water Qual 1991, 6, 177᎐184.
Harada, K.-I. Toxic Microcystis; CRC Press: Boca Raton, FL,
1996; pp 103᎐148.
Helenius, A.; Simons, K. Biochim Biophys Acta 1975, 415,
2
9᎐79.
H ¨o fte, H.; Whiteley, H. R. Microbiol Rev 1989, 53, 242᎐255.
Honkanen, R. E.; Zwiller, J.; Moore, R. E.; Daily, S. L.;
Khatra, B. S.; Dukelow, M.; Boynton, A. L. J Biol Chem
qqq: annihilation, qq": almost annihilation, qq: half of
the larvae dead, q: some larvae dead, ": unclear activity, ᎏ: no
activity.
1
990, 265, 19410᎐19404.
Ikawa, M.; Haney, J. F.; Sasner, J. J Hydrobiol 1996, 331,
67᎐170.
Kiviranta, J.; Abdel-Hameed, A. World J Microbiol Biotech
994, 10, 517᎐520.
1
998; Ikawa et al., 1996; Morohashi et al. 1991.. Based
on the results obtained from the present and previous
studies, the activity of unsaturated fatty acids to
mosquito larvae may be derived from the inhibition of
Na rK -ATPase. Similarly, brine shrimp, Artemia
salina, would be affected by the unsaturated fatty acids
in cyanobacteria, because it showed inhibitory behavior
similar to mosquito larvae ŽKiviranta and Abdel-
Hameed, 1994; Morohashi et al., 1991..
1
1
q
q
Kiviranta, J.; Sivonen, K.; Niemel ¨a , S. I.; Huovinen, K. Envi-
ron Toxicol Water Qual 1991, 6, 423᎐436.
Lange, L; Lopez, C. S. Protection Agents from Nature: Natu-
ral Products and Analogues; Royal Society of Chemistry:
Cambridge, UK, 1996; Chap 1.
Lawton, L. A.; Beattie, K. A.; Hawser, S. P.; Campbell, D. L.;
Codd, G. A. Detection Methods for Cyanobacterial Toxins;
The Royal Society of Chemistry: Cambridge, UK, 1994; pp
Several invertebrates such as brine shrimp, mosquito,
and Daphnia have been investigated for use in routine
bioassays for cyanobacterial toxins such as microcystins
and anatoxins ŽKiviranta et al., 1991; Lawton et al.,
1
11᎐116.
MacKintosh, C.; Beattie, K. A.; Klumpp, S.; Cohen, P.; Codd,
G. A. FEBS Lett 1990, 264, 187᎐192.
Moore, R. E.; Corbett, T. H.; Patterson, G. M. L; Valeriote,
F. A. Curr Pharm Des 1996, 2, 317᎐330.
Morohashi, M.; Tsuchiya, K.; Mita, T.; Kawamura, M. J
Comp Physiol B 1991, 161, 69᎐72.
Mulla, M. S.; Mian, L. S.; Kawecki, J. A. Residue Reviews;
Springer-Verlag: New York, 1981; pp 1᎐136.
1
994.. Of these, the brine shrimp Ž Artemia salina. has
been the most popular and is available in commercial
kit. In this study such invertebrates were affected by
nontoxic cyanobacterium, suggesting that the observed
toxicity of toxic cyanobacterium is derived not only
from cyanobacterial toxins but also from unsaturated
fatty acids, when a bioassay with invertebrates was
used.
After the bioassay method was developed for these
hydrophobic compounds, we further demonstrated that
unsaturated fatty acids from cyanobacterium also have
an activity against mosquito larvae. At present it is
impossible to establish whether nontoxic cyanobacteria
could be used as an origin for safe insecticides which
should not cause serious damage to the environment.
We are now focusing on another activity of the water-
soluble fraction from this cyanobacterium.
Murakami, N.; Morimoto, T.; Imamura, H.; Ueda, T.; Nagai,
S.; Sakakibara, J.; Yamada, N. Chem Pharm Bull 1991, 39,
2
277᎐2281.
Namikoshi, M; Rinehart, K. L. J Ind Microbiol 1996, 17,
73᎐384.
3
Nishiwaki-Matsushima R.; Ohta, T.; Nishiwaki, S.; Suganuma,
M.; Kohyama, K.; Ishikawa, T.; Carmichael, W. W.; Fujiki,
H. J Cancer Res Clin Oncol 1992, 118, 420᎐424.
Ohta, T.; Sueoka, E.; Iida, N.; Komori, A.; Suganuma, M.;
Nishiwaki, R.; Tatematsu, M.; Kim, S.-J.; Carmichael, W.
W.; Fujiki, H. Cancer Res 1994, 54, 6402᎐6406.
We thank Dr. Jussi Meriluoto, Department of Biochem-
istry and Pharmacy, A¨ bo Akademi University, Finland, for
Rinehart, K. L.; Namikoshi, M.; Choi, B. W. J Appl Phycol
1994, 6, 159᎐176.
helpful discussions.
Sivonen, K.; Jones, G. Toxic Cyanobacteria in Water, E & FN
Spon: London, 1999; pp 41᎐111.
Tanaka, K; Mizusawa, K; Saugstad, E. S. Contrib of the Am
Entomol Inst 1979, 16, 1᎐987.
REFERENCES
Yoshizawa, S.; Matsushima, R.; Watanabe, M. F.; Harada,
K.-I.; Ichihara, A.; Carmichael, W. W.; Fujiki, H. J Cancer
Res Clin Oncol 1990, 116, 609᎐614.
Bury, N. R.; Codd, G. A.; Wendelaar Bonga, S. E.; Flik, G.
J Exp Biol 1998, 201, 81᎐89.