Stereostructure of komodoquinone A, a neuritogenic anthracycline, from marine Streptomyces sp. KS3

Article abstract of DOI:10.1248/cpb.51.1402

The absolute stereostructure of komodoquinone A (1), a neuritogenic anthracycline, which was isolated from a cultured marine Streptomyces sp. KS3, has been determined on the basis of chemical derivatization. Komodoquinone A (1) induces neuronal cell differentiation in the neuroblastoma cell line, Neuro 2A and arrests the cell cycle at the G1 phase. The effect of a solid-state medium on the production of 1 and its aglycone, komodoquinone B (2), was examined.

Full text of DOI:10.1248/cpb.51.1402

1
402
Chem. Pharm. Bull. 51(12) 1402—1404 (2003)
Vol. 51, No. 12
Stereostructure of Komodoquinone A, a Neuritogenic Anthracycline, from
Marine Streptomyces sp. KS3
Takuya ITOH, Masahiro KINOSHITA, Hong WEI, and Motomasa KOBAYASHI*
Graduate School of Pharmaceutical Sciences, Osaka University; 1–6 Yamada-oka, Suita, Osaka 565–0871, Japan.
Received August 1, 2003; accepted September 24, 2003
The absolute stereostructure of komodoquinone A (1), a neuritogenic anthracycline, which was isolated
from a cultured marine Streptomyces sp. KS3, has been determined on the basis of chemical derivatization. Ko-
modoquinone A (1) induces neuronal cell differentiation in the neuroblastoma cell line, Neuro 2A and arrests the
cell cycle at the G1 phase. The effect of a solid-state medium on the production of 1 and its aglycone, komodo-
quinone B (2), was examined.
Key words komodoquinone A; anthracycline; Neuro 2A; marine; Streptomyces sp.; neuritogenic activity
1
3)
A growing number of biologically active metabolites have production of secondary metabolites. Komodoquinones A
1
4)
been explored from marine microorganisms during the past (1) and B (2) together with staurosporine were produced by
two decades. In recent years, marine microorganisms are a marine Streptomyces sp. KS3 cultured in a solid-state
paid much attention as a prodigious source for novel drug medium based on rice. In the case that the KS3 strain was
1
—4)
discovery.
In the course of our project searching for new grown in a ISP1ϩglucose liquid medium, no production of 1
5
,6)
bioactive substances from marine organisms, we found a and 2 was observed. These results led us to investigate the
novel anthracycline named komodoquinone A (1) and its
aglycone, komodoquinone B (2), from the solid-state fer-
mentation (solid medium based on rice) of a marine Strepto-
7
)
myces sp. KS3. Komodoquinones A (1) and B (2) are char-
acterized as a rare anthracycline having a 9-methyl sub-
8
)
9)
stituent (e.g. nogalamycin, steffinomycin, and mutacti-
1
0)
mycin ). The absolute stereo-structure of komodoquinone B
2) with 7,9-syn-dihydroxyl groups has been confirmed by
application of the modified Mosher’s method. The amino
sugar portion in 1 was also characterized to be linked to the
hydroxyl group in the D-ring of anthracyclinone. The NMR
analysis of the amino sugar portion elucidated that this
(
1
1)
1
2)
amino sugar is the C-4 stereo-isomer of mycaminose,
which has been found as a component in macrolide antibi-
otics, and a novel type of amino sugar. This paper presents
the elucidation of the absolute structure of the sugar portion
in komodoquinones A (1), the effect of a solid-state medium Fig. 1. Structure of Komodoquinones A (1) and B (2)
on the production of 1 and 2, and the neuritogenic activity of
1
against Neuro 2A cells.
Results and Discussion
In order to determine the absolute stereo-structure of the
sugar portion in 1, the following chemical conversion was
executed. Compound 1 was subjected to NaIO degradation
4
followed by acid treatment (2 N aq. HCl at 50 °C) to give a
C3-aldehyde, which was further treated with phenylhy-
drazine and acetic acid to furnish 1-(phenylhydrazo)propan-
2
-ol (4c). The authentic hydrazone derivatives 4a and 4b
were also prepared from methyl L-rhamnoside (3) and methyl
D-fucoside (5) applying the same synthetic route (Fig. 2).
HPLC analysis of these hydrazone derivatives clarified that
4
c was identical with 4a. Therefore, the amino sugar moiety
in 1 has been defined to be L-form and the absolute stereo-
structure of komodoquinone A was confirmed to be 1 (Fig.
1
).
Solid-state fermentation is well known to be used for pro-
duction of brewages, sake, and soy-sauce. In recent years,
solid-state fermentation has been applied successfully for Methyl L-Rhamnoside (3) and Methyl D-Fucoside (5)
Fig. 2. Hydrazone Derivatives Prepared from Komodoquinone A (1),
∗
To whom correspondence should be addressed. e-mail: kobayasi@phs.osaka-u.ac.jp
© 2003 Pharmaceutical Society of Japan

December 2003
1403
role of the solid-state fermentation on the growth of the KS3
strain and its ability to produce the secondary metabolites.
We examined the effect of other cereals such as corn, soy-
bean, and barley, on the production of 1 and 2. The solid-
state medium based on corn or soybean was not able to sup-
port the growth of the KS strain. However, in the case of ei-
ther barley or rice, the strain grew very well and produced
komodoqinones A (1) and B (2). On the other hand, the culti-
vation of the KS3 strain in ISP1ϩglucose medium-added
rice produced only komodoquinone B (2). Therefore, it was
presumed that a constituent contained in barley or rice was
concerned with the biosynthesis of the aglycone, komodo-
quinone B (2), in the fermentation of the KS3 strain. How-
ever, the biosynthesis and glycosidation of the amino sugar
portion in 1 must be associated with the contribution of some
parameter such as osmotic pressure, which usually exert a
substantial effect in solid-state fermentation. Consequently,
solid-state fermentation of microorganisms is expected to be
a good strategy to explore novel bioactive compounds.
Many anthracyclines have been reported from terrestrial
actinomycetes, which have been documented to exhibit anti-
1
5)
microbial, anti-tumor and other biological activities. How-
ever, komodoquinone A (1) is the first instance of anthracy-
cline exhibiting neuritogenic activity. Compound 1 was
1
6,17)
shown to induce a morphological change
with multipolar
processes emanating from the body of Neuro 2A cells at a
concentration of 1 mg/ml (Fig. 3). In contrast, adriamycin (6),
a repesentative anthracycline, showed no neuritogenic activ-
ity even at the same concentration, and komodoquinone B
(
2), the aglycone of 1, showed only weak activity at a con-
7
)
centration of 30 mg/ml. These results imply that the amino
sugar moiety bound to the D-ring in 1 might play an impor- Induced Morphological Changes in Neuro 2A Cells
tant role for neuritogenic activity.
Next, we investigated the effect of 1 and 6 on the cell cycle
of Neuro 2A cells, since neuronal differentiation was closely
related to the cell cycle. The cell cycle of Neuro 2A cells
treated with 1 (at 1 mg/ml) for 24 h was arrested at the G1
phase, while that treated with 6 (at 1 mg/ml) for 24 h was ar-
rested at the G2/M phase (Fig. 4). Generally, anthracycline
antibiotics, which intercalate in DNA and RNA, have been
known to arrest the cell cycle at the G2/M phase. There-
fore, komodoquinone A (1) may exert neuronal differential
activity by a mechanism, which is different from the interca-
lation in DNA.
Fig. 3. Photomicrographs of Komodoquinone A (1) and Adriamycin (6)-
Cells were cultured in the presence of 1 (1 mg/ml; a) or 6 (1 mg/ml; b) or absence
(
control; c) for 24 h.
1
8)
1
9)
Experimental
The following instruments were used to obtain physical data: a Jasco DIP-
1
81 digital polarimeter for specific rotations; a Hitachi 260-30 infrared spec-
trometer for IR spectra; a Varian Inova-600 (600 MHz) NMR spectrometer
1
13
for H- and C-NMR (150 MHz) spectra. Silica gel (Fuji silysia BW-200)
and pre-coated thin layer chromatography (TLC) plates (Merck, Kiesel gel,
6
0F254) were used for column chromatography and TLC. Spots on TLC
plates were detected by spraying vanillin/H SO (vanillin 5 g, c-H SO
2
4
2
4
9
2
5 ml) or acidic p-anisaldehyde solution (p-anisaldehyde 25 ml, c-H SO4
5 ml, acetic acid 5 ml, EtOH 425 ml) with subsequent heating.
Preparation of Hydorazone Derivatives 4a—c A solution of methyl L-
2
Fig. 4. The action of Komodoquinone (1) and Adriamycin (6) on the Cell
Cycle of Neuro 2A Cells
rhamnoside (3, 80 mg) in MeOH–water (1 : 1, 5 ml) was treated with NaIO4
Neuro 2A cells were exposed to 1 or 6 for 24 h, and the cell cycle distribution was
determined by flow cytometric analysis.
(
1.4 g) for 12 h at room temperature, and then the reaction mixture was ex-
tracted with AcOEt. The AcOEt layer was washed with brine and dried over
MgSO . Removal of solvent from the AcOEt layer under reduced pressure
4
gave a corresponding aldehyde. A solution of the aldehyde in 2 N HCl
dried over MgSO and evaporated in vacuo. The crude product was purified
4
(
10 ml) was stirred for 1 h at 50 °C. After neutralization by Na CO , the re- by reversed phase HPLC [Cosmosil 5C18-AR (10 mm i.d.ϫ250 mm),
2 3
action mixture was added to phenylhydrazine (100 ml) and a drop of acetic
mobile phase; MeOH–waterϭ7 : 3, detection; UV (lϭ268 nm), flow rate;
acid. After further 1 h stirring, the reaction mixture was quenched with 2.0 ml/ml] to afford 1R-(phenylhydrazo)propan-2-ol 4a (11 mg). A solution
brine, then the whole was extracted with AcOEt. The AcOEt phase was of methyl D-fucoside (5, 80 mg) was similarly treated to obtain 1S-(phenyl-

1
404
Vol. 51, No. 12
4
hydrazo)propan-2-ol 4b (8 mg).
Cell Cycle Analysis The cell suspension of Neuro 2A cells (4ϫ10
A solution of komodoquinone A (1, 4.0 mg) in MeOH–water (1 : 1, cells in 1 ml of the culture medium) was plated on 24-well plate and incu-
0
.1 ml) was treated with NaIO (10 mg) for 12 h at room temperature, and bated for 24 h. After medium exchange, an ethanol solution (10 ml) of the
4
then the reaction mixture was extracted with AcOEt. The AcOEt layer was test sample was added and further incubated for 24 h. The culture medium of
washed with brine and dried over MgSO . Removal of solvent from the the cell suspension was removed by centrifugation (1000 g for 3 min). The
4
AcOEt layer under reduced pressure gave a corresponding aldehyde. A solu- collected cells were dyed by DNA-Prep Reagents Kit for 20 min. Then, the
tion of the aldehyde in 2 N HCl (0.2 ml) was stirred for 1 h at 50 °C. The re- supernatant was removed by centrifugation (500 g for 5 min) and the result-
action mixture was similarly treated to afford 1-(phenylhydrazo)propan-2-ol ing cell suspension in 500 ml of D-PBS (Ϫ) solution was filtered by 40-mm
4
c.
Hydrazone Derivative 4a: [a]D Ϫ53° (cϭ0.2, MeOH). IR (KBr) cm
nylon mesh filter. The cell cycle analysis of the filtrate was carried out on a
Ϫ1
:
FACSCalibur (Becton Dickinson, l ϭ493 nm, lemϭ630 nm).
ex
1
3
345, 1653, 1585. UV lmax (MeOH) nm (e): 268 (8600). H-NMR
(
600 MHz, C D ) d: 7.2—7.0 (7H, m, –Ph, –NH, –CNH), 4.15 (1H, q,
Acknowledgments The authors are grateful to the Houansha Founda-
6
6
Jϭ6.6 Hz, 2-H), 3.87 (1H, s, 2-OH), 1.11 (3H, d, Jϭ6.6 Hz, 3-H). FAB-MS: tion, the Tokyo Biochemical research foundation, and the Uehara Memorial
ϩ
ϩ
m/z 165 (MϩH ). High resolution FAB-MS: m/z 165.1033 (MϩH , Calcd Foundation for financial support.
for C H N O: 165.1028).
9
12
2
Ϫ1
Hydrazone Derivative 4b: [a]D ϩ52° (cϭ0.2, MeOH). IR (KBr) cm
:
References
1
3
345, 1655, 1580. UV lmax (MeOH) nm (e): 268 (8600). H-NMR
1) Jensen P. R., Fenical W., “Drug from the Sea,” ed. by Fusetani N.,
Karger, Basel, 2000, pp. 6—29.
(
600 MHz, C D , d): 7.2—7.0 (7H, m, –Ph, –NH, –CNH), 4.15 (1H, q,
6
6
Jϭ6.6 Hz, 2-H), 3.85 (1H, s, 2-OH), 1.11 (3H, d, Jϭ6.6 Hz, 3-H). FAB-MS:
2) Jensen P. R., Fenical W., Annu. Rev. Microbiol., 48, 559—548 (1994).
3) Jensen P. R., Fenical W., J. Ind. Microbiol., 17, 346—351 (1996).
4) Pietra F., Nat. Prod. Rep., 14, 453—464 (1997).
5) Kobayashi M., Higuchi K., Murakami N., Tajima H., Aoki S., Tetrahe-
dron Lett., 38, 2859—2862 (1997).
6) Aoki S., Yoshioka Y., Miyamoto Y., Higuchi K., Setiawan A., Mu-
rakami N., Chen Z., Sumizawa T., Akiyama S., Kobayashi M., Tetra-
hedron Lett., 39, 6303—6306 (1998).
7) Itoh T., Kinoshita M., Aoki S., Kobayashi M., J. Nat. Prod., 66,
1373—1377 (2003).
8) Wiley P. F., Kelly R. B., Caron E. L., Wiley V. H., Johnson J. H.,
MacKellar F. A., Mizsak S. A., J. Am. Chem. Soc., 99, 542—549
(1997).
ϩ
ϩ
m/z 165 (MϩH) . High resolution FAB-MS: m/z 165.1031 (MϩH , Calcd
for C H N O: 165.1028).
Chiral HPLC Analysis of Hydrazone Derivatives 4a—c Hydrazone
derivatives 4a—c were analyzed by chiral HPLC [CHIRALCEL OD-R,
9
12
2
4
.6 mm i.d.ϫ250 mm, mobile phase; CH CN–H Oϭ3 : 7, detection; UV
3
2
(
268 nm), flow rate; 0.5 ml/min]. The retention times of 4a—c were 34.5,
3
0.0, and 34.5 min, respectively.
Culture of Streptomyces sp. KS3 in Other Conditions A liter of seed
medium (ISP1ϩglucose) [5 g of Bacto triptone, 3 g of yeast extract, 5 g of
glucose, and artificial sea water] was autoclaved before use. The rice solid
medium [25 g of rice, 200 mg of glucose, 100 mg yeast extract, and 50 ml of
artificial sea water] in 500 ml flask was autoclaved before use. The barley
solid medium [15 g of barley, 200 mg of glucose, 100 mg yeast extract, and
9) Kunnari T., Tuikkanen J., Hautala A., Hakala J., Ylihonko K.,
Mantsala P., J. Antibiot., 50, 496—501 (1997).
3
0 ml of artificial sea water] in 500 ml flask was autoclaved before use. The
soybean solid medium [25 g of soybean, 200 mg of glucose, 100 mg yeast 10) Speitlin M., Nattewan P., Yazawa K., Mikami Y., Wollny I., Ritzau M.,
extract, and 40 ml of artificial sea water] in 500 ml flask was autoclaved be-
fore use. The corn solid medium [25 g of corn, 200 mg of glucose, 100 mg
yeast extract, and 40 ml of artificial sea water] in 500 ml flask was auto-
Laatsch H., Grafe U., J. Antibiot., 51, 693—698 (1998).
11) Ohtani I., Kusumi T., Kashman Y., Kakisawa H., J. Am. Chem. Soc.,
113, 4092—4096 (1991).
claved before use. A liter of ISP1ϩglucose-rice medium [25 g of rice, 5 g of 12) Omura S., Nakagawa A., Neszmelyi A., Gero S. D., Sepulchre A.,
Bacto triptone, 3 g of yeast extract, 5 g of glucose, and artificial sea water]
Piriou F., Lukacs G., J. Am. Chem. Soc., 97, 4001—4009 (1975).
was autoclaved before use. The Streptomyces sp. KS3 starin was cultured in 13) Robinson T., Singh D., Nigam P., Appl. Microbiol. Biotechnol., 55,
the seed medium at 30 °C for 5 d. Then, the broth of the strain was inocu-
4284—4289 (2001).
lated into the production medium and cultured under static conditions at 14) Martinsson T., Fowler C. J., Life Sci., 53, 1557—1565 (1993).
0 °C for 2 weeks.
15) Demain A. L., Appl. Microbiol. Biotechnol., 52, 455—463 (1999).
Assay for Neuritogenic Activity in Neuro 2A Cells Neuro 2A cells 16) Blanco V., Camelo J. L., Carri N. G., Cell Biol. Internat., 25, 909—
were grown in Dulbecco’s modified essential medium (DMEM) with 10%
917 (2001).
fetal bovine serum (FBS). The cells were kept in an incubator at 37 °C with 17) Fenteany G., Schreiber S. L., J. Biol. Chem., 273, 8545—8548 (1998).
3
4
5
% CO . The cells were plated on 24-well plates at a density of 2ϫ10 per 18) Aoki S., Matsui K., Tanaka T., Wei H., Kobayashi M., Biochem. Bio-
2
well with 1 ml of culture medium. After 24 h cultivation, the medium was
phys. Res. Commun., 289, 558—563 (2001).
exchanged for fresh medium, and the testing sample as 10 ml of EtOH solu- 19) Kimler B. F., Schneiderman M. H., Leeper D. B., Cancer Res., 38,
tion was added to each well. After 24 or 48 h incubation, morphological
changes in the cells were observed under microscope.
809—814 (1978).