Mechanistic studies on the biomimetic reduction of tetrahydroxynaphthalene, a key intermediate in melanin biosynthesis

Article abstract of DOI:10.1248/cpb.49.192

1,3,6,8-Tetrahydroxynaphthalene (T4HN) is an aromatic polyketide, serving as a general precursor of fungal melanin. Melanin biosynthesis involves two consecutive deoxygenations of T4HN, consisting of the reduction of a phenolic carbon followed by dehydrat

Full text of DOI:10.1248/cpb.49.192

192
Chem. Pharm. Bull. 49(2) 192—196 (2001)
Vol. 49, No. 2
Mechanistic Studies on the Biomimetic Reduction of
Tetrahydroxynaphthalene, a Key Intermediate in Melanin Biosynthesis1)
,2)
*
*
Koji ICHINOSE, Yutaka EBIZUKA, and Ushio SANKAWA
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113–0033, Japan.
Received September 13, 2000; accepted November 6, 2000
1,3,6,8-Tetrahydroxynaphthalene (T4HN) is an aromatic polyketide, serving as a general precursor of fungal
melanin. Melanin biosynthesis involves two consecutive deoxygenations of T4HN, consisting of the reduction of a
phenolic carbon followed by dehydration. The first reduction to produce scytalone was studied in a biomimetic
reduction with sodium borohydride. The reduction required a strong alkaline condition, leading to the tautomer-
ization of T4HN to a reactive species whose structure was clarified by NMR spectroscopy.
Key words 1,3,6,8-tetrahydroxynaphthalene; melanin biosynthesis; scytalone; biomimetic; polyketide
mimetic reduction15) of 2 to 3 to obtain further chemical in-
Melanin biosynthesis in a number of pathogenic fungi is
formation leading to better understanding of post-aromatic
deoxygenation.
an essential step for subsequent infectious procedures such
as the formation of appressorium.3) Fungal melanin is de-
rived from the polymerization of 1,8-dihydroxynaphthalene
(DHN, 1), which is biosynthesized from a pentaketide inter-
mediate, 1,3,6,8-tetrahydroxynapthalene (T4HN, 2) by a se-
ries of reductions and dehydrations involving scytalone (3),
1,3,8-trihydroxynaphthalene (T3HN, 4) and vermelone (5)
(Chart 1).3) Biosynthesis of 2 is catalyzed by a polyketide
synthase (PKS).4) Fungal PKS is a multifunctional large pro-
tein consisting of catalytic domains such as b-ketoacyl syn-
thase (KS), acyl transferase (AT), acyl carrier protein (ACP),
dehydratase (DH), enoyl reductase (ER), ketoreductase (KR)
and thioesterase (TE).4) The PKS1 gene was cloned5) from
the phytopathogenic fungus, Colletotrichum lagenarium, by
complementation experiments of melanin-deficient albino
mutants of C. lagenarium. The deduced gene product carries
KS, AT, ACP and TE domains, presumed to encode a T4HN
synthase. Recently, expression of the PKS1 gene in a heterol-
ogous Aspergillus oryzae, demonstrated6) that the enzymatic
product of the PKS1 is indeed T4HN (2).
The reduction step of T4HN (2) is of great interest mecha-
nistically, since it is, together with the subsequent dehydra-
tion, an example of post-aromatic deoxygenation of acetate-
derived oxygen in the biosynthesis of aromatic polyketides of
microbial origin. Other examples are found in the biosynthe-
sis of chrysophanol7,8) in Pyrenochaeta terrestris, and of
sterigmatocystin9) in Aspergillus parasisticus. From the
chemical point of view, the reduction occurring at a phenolic
carbon is rather unusual in contrast to the loss of an oxygen
atom before aromatization, as seen in the biosynthesis of 6-
methylsalicylic acid.10) In our earlier biosynthetic studies11)
on 3, [1,2-13C2]-acetate was incorporated into 3 with two dif-
ferent labeling patterns in a 1 : 1 ratio, indicating that 3 is
biosynthesized via a synmetrical intermediate, 2. It was then
demonstrated12) that 2 was reduced to 3 by cell-free extracts
of Verticillium dahliae,12) as well as of Phialophora lager-
bergii8) in the presence of NADPH. Recent genetic studies on
fungal melanin biosynthesis identified the genes encoding
polyhydroxynaphthalene reductase: ThnR13) from Magna-
porthe grisea; THR114) from C. lagenarium. It is also note-
worthy that some non-fungicidal anti-blast chemicals, such
as tricyclazole, selectively inhibit the reductions.3) In the pre-
sent study we have conducted mechanistic studies on a bio-
Results and Discussion
Melanin-precursors, polyhydroxynapththalenes, readily
undergo air-oxidation to form quinonic compounds such as
flaviolin (6) and 2-hydroxyjuglone (7), together with poly-
merized pigments (Chart 1).3) Compound 2 is particularly la-
bile and was previously characterized as its stable tetraac-
etate.6,15) The present study required the development of a
practical method for the preparation of 2. A modification
(see Experimental) of the method15) of Tanaka et al., which
use alkali fusion of sodium chromotropic acid (12), resulted
in a significant improvement (58%) (Chart 2).
Bycroft et al.16) reported that 2 was prepared from methyl
curvulinate (11) in a sodium methoxide solution, where in
situ reduction with sodium borhydride proceeded to afford 3.
This led us to try a reduction of 2, which had been prepared
separately, with sodium borohydride in methanol, which
turned out to be unsuccessful. The reported method, there-
fore, was reinvestigated. Compound 11 was synthesized from
commercially available 8, as shown in Chart 2. Treatment of
11 with sodium methoxide, followed by the addition of
sodium borohydride, gave
3 (20%) 6,8-dihydroxy-1-
methylisochroman-3-one (5%, 13), along with quinonic com-
Chart 1. Biosynthesis of Fungal Melanin and Related Metabolites Derived
from 1,3,6,8-Tetrahydroxynaphthalene (T4HN, 2)
To whom correspondence should be addressed. e-mail: yebiz@mol.f.u-tokyo.ac.jp
© 2001 Pharmaceutical Society of Japan

February 2001
193
Chart 2. Synthesis of T4HN and Related Materials Used for the Present Study
a) BzBr, K2CO3 (96%); b) LiAlH4 (98%); c) CH3SO2Cl, NEt3 (84%); d) NaCN, BzNEt3Cl (69%); e) Ba(OH)2 (92%); f) CH2N2 (100%); g) Ac2O–60% HClO4 aq. (99%) for 11;
h) [1,2-13C2] AcOH-(CF3CO)2CO (89%) for 11a; i) Pd/C (97%); j) NaOH–Ba(OH)2 (58%).
1
acetylation of 10 with [1,2-13C2] acetic acid. The H-decou-
pled 13C-NMR spectrum (Fig. 3) of the cyclized 11a gave
two pairs of 13C–13C coupled signals, d: 102.1 and 192.6
(Jϭ63 Hz) and d: 105.3 and 165.0 (Jϭ68 Hz), clearly indi-
cating that the C-3 carbon was assignable to the upper-field
1
signal (192.2 ppm). Thus, complete assignments of the H-
and 13C-NMR signals were made (Table 1).
In conclusion, the reactive species is a keto-tautomer rep-
resented by 14a and 14b, being readily formed from 2 under
strong alkaline conditions. The fact that 3 is formed as a sin-
Chart 3. Reaction of T4HN (2) with NaBH4 in CH3ONa–CH3OH
gle product in the biomimetic reduction of 2 suggests 14b to
pounds formed by air oxidation (Chart 3). Apparently, the be an active species for the reduction at C-3 with sodium
presence of sodium methoxide is a key factor for the success- borohydride. This was rigorously supported by our theoreti-
ful reduction. Sodium borohydride reduction of 2 in cal calculations: 2 forms a stable keto-tautomer trianion rep-
methanol was conducted in the presence of sodium methox- resented by 14a or 14b under alkaline conditions induced by
ide, and 3 was obtained in 25% yield, along with the sodium methoxide;17) a hydride attack at C-3 of the trianionic
quinonic compounds (Chart 3). The results indicate that 2 species is more favorable than at the other possible reactive
was converted into its reactive species only under a strong al- position, C-1.18)
kaline condition. We then tried to identify the reactive
species using NMR spectroscopy.
A plausible mechanism of the enzymatic reduction of
T4HN involves its efficient deprotonation steps in the en-
Compound 11 was treated with a deuteromethanol zyme complex to generate a reactive keto-tautomer that sub-
(CD3OD) solution containing sodium deuteromethoxide sequently undergoes hydride transfer from NADPH. Under
(CD3OD) under Ar, and the reaction mixture was directly neutral conditions, NMR spectral profiles of T4HN (2) in an
submitted to NMR measurement. Separately, NMR spectra aprotic solvent (acetone-d6), and in a protic one (CD3OD) are
of 2 were measured in a CD3ONa–CD3OD solution. The 13C- significantly different (see Experimental). The former gave
NMR spectra of the two samples were essentially identical. signals solely derived from a symmetrical naphthalene form,
1
The H-decoupled 13C-NMR spectra showed ten significant but the latter provided a mixture of naphthalene (87.5%) and
signals, indicating that the reactive species is not a symmetri- keto-tautomer (12.5%) species. Extensive NMR analysis es-
cal compound (Fig. 1a). Two signals at d: 192.2 and 192.6, tablished full assignments of the signals derived from the
were attributed to conjugated carbonyl and enol carbons, re- keto-tautomer. The key signal of C-3 at 178.0 ppm is shifted
spectively. Four broad signals at d: 41.0, 102.1, 105.3 and to upfield by ca. 14 ppm from that in strong alkali, suggest-
112.2 were assignable to carbon atoms bearing deuterium, ing the C-3 carbon is less reactive to the reducing reagent.
1
which were formed by H–2H exchange. This is consistent Sodium borohydride reduction in the presence of CH3ONa
with the following data: (i) 1H-NMR monitoring experiments was also attempted for the second substrate, 4, but no re-
of the cyclization of 11 under stepwise elevations in tempera- duced product was obtained. The 13C-NMR spectrum of 4 in
ture, where signals derived from the product disappear as the CD3ONa–CD3OD gave ten signals, all assignable to a naph-
1
2
reaction proceeds (Fig. 2); (ii) The H, H-decoupled 13C- thalene type species (see Experimental), showing that 4 dis-
NMR spectrum showed that respective signals changed into favors the formation of a reactive keto-tautomer under the
sharp singlets (Fig. 1b).
To assign all the NMR signals, we measured 2D-NMR
present alakaline conditons.
Previous biochemical studies8,13) suggested that the enzy-
spectra using CD3OD–CH3OH (1 : 1) as a solvent. Three sin- matic reductions of 2 and 4 are catalyzed by the same en-
1
glet H signals were observed at d: 5.16, 5.88 and 6.02, zyme. The present results, however, indicate that the chemi-
which correlated with the signals at d: 102.1, 105.3 and cal properties of 2 and 4 are significantly different, present-
112.2 in a C–H COSY spectrum, respectively. The HMBC ing a very interesting biochemical problem: how a single en-
3
spectrum revealed key correlations based on JC–H involving zyme accommodates the substrate, either 2 or 4, in the cat-
quarternary carbons (Table 1). A definite assignment for the alytic pocket in such a way that an efficient hydride transfer
closely observed signals at d: 192.2 and 192.6 was made occurs at the C-3 carbon.
using a doubly 13C-labeled T4HN. Methyl curvulinate doubly
labeled at the acetyl moiety (11a) was synthesized by the

194
Vol. 49, No. 2
Fig. 1. 13C-NMR Spectra of T4HN (2) in CH3ONa–CH3OH: (a) 1H-Decoupled; (b) 1H, 2H-Decoupled
Fig. 2. Changes of 1H-NMR Spectra during the Formation of T4HN (2) from Methyl Curvulinate (11)
Fig. 3. 1H-Decoupled 13C-NMR Spectrum of T4HN Prepared from 13C-Labeled Methyl Curvulinate (11a)

February 2001
195
Table 1. NMR Spectral Data for T4HN (2) in CD3ONa–CD3OD
as colorless prisms (mp 82—83 °C). The product in ethanol was treated with
5% Pd/C under H2 to afford 11 (0.47 g, 97%) as a white powder. mp 135—
136 °C. Anal. Calcd for C11H12O5: C, 58.92; H, 5.40. Found: C, 58.63; H,
Carbons correlated in
HMBC spectruma)
Position
d
C (ppm)
d
H (ppm)a)
1
5.41. EI-MS m/z: 224 (Mϩ), 182 (MϪAc). H-NMR (400 MHz, acetone-d6)
d: 2.53 (3H, s, CH3CO–), 3.63 (3H, s, OCH3), 3.79 (2H, s, –CH2COOCH3),
6.33 (1H, d, Jϭ2.3 Hz, 4-H), 6.37 (1H, d, Jϭ2.3 Hz, 6-H), 9—10 (br –OH).
13C-NMR (100 MHz, acetone-d6) d: 32.6, 40.9, 52.4, 103.2, 112.8, 119.2,
138.3, 161.9, 162.2, 172.6, 204.0.
C-1
C-2
C-3
C-4
C-4a
C-5
C-6
C-7
C-8
C-8a
192.6
102.1
192.2
41.0
143.4
112.2
174.5
105.3
165.0
107.7
—
5.16
—
4, 8a
—
—
—
4, 6, 7, 8a
—
5, 6, 8, 8a
—
—
13C-Doubly Labeled Methyl Curvulinate (11a) 868 mg of 10 was dis-
solved in a mixture of [1,2-13C2] acetic acid (99.7 atom % 13C) (0.1 ml) and
acetic acid (0.1 ml). Trifluoroacetic anhydride (0.6 ml) was added to the mix-
ture at 4 °C, and it was then stirred for 45 min at room temperature, poured
into saturated aqueous NaHCO3 and extracted with ethyl acetate. The usual
work-up and crystallization from benzene–hexane gave a dibenzylether of
the product (863 mg, 89%). Enrichment of 13C was evaluated to be 51%
based on the inverse gated decoupled 13C-NMR spectrum. 11a was prepared
by quantitative hydrogenation of the dibenzylether with Pd/C, as described
above.
b)
—
6.02
—
5.88
—
—
—
a) Measured in a CD3OD–CH3OH (1 : 1) solution of sodium methoxide. b) Over-
lapped with methanol signal.
1,3,6,8-Tetrahydroxynaphthalene (T4HN) (2) Sodium chromotropic
acid (12 g) was heated at 260 °C for 7 h under N2 in the presence of NaOH
(48 g), Ba(OH)28H2O (72 g), and KOH (72 g). After cooling under reduced
Experimental
Melting points were determined on a Yanagimoto micro melting point ap- pressure, the resultant solid was triturated with 10% H2SO4 to give an aque-
paratus and uncorrected. Column chromatography was carried out with ous suspension, followed by extraction with ether. The concentrated extract
Wakogel C-200. TLC was conducted on a 0.25 or 2 mm precoated silica gel was immediately subjected to flash chromatography on silica gel with
plate (60F254, Merck). NMR and MS spectra were measured on JEOL GSX- ether–hexane (4 : 1). The eluent containing the product was concentrated to
400 or 500 and JEOL JMS DX-300, respectively. [1,2-13C2] Acetic acid was precipitate 2 (3.7 g, 58%) as a yellowish powder. EI-MS m/z: 194 (Mϩ).
1
purchased from MSD Isotopes Co., Ltd.
HRMS m/z: 192.0404 (Calcd for C10H8O4: 192.0404). H-NMR (400 MHz,
(3؅,5؅-Dibenzyloxy)phenylacetonitrile (9) Potassium carbonate (103 g) acetone-d6) d: 6.23 and 6.45 (each 2H, d, Jϭ2.1, 2.0 Hz, 2, 7-H and 4, 5-H).
1
and benzyl bromide (75 ml) were added to an anhydrous acetone (400 ml) The H-NMR spectrum of 2 in CD3OD indicates the presence of a mixture
solution of methyl 3,5-dihydroxybenzoate (8) (50.4 g). The mixture was re- of naphthalene type (87.5%) and keto-tautomer type (12.5%) under neutral
fluxed for 12 h. After cooling, the precipitate was removed by filtration and conditions. Extensive NMR studies established the following assignments of
the filtrate was evaporated. The residue was crystallized from hexane to give both species.
methyl 3,5-dibenzyloxybenzoate (100.4 g, 96%) as a white powder (mp
67—68 °C). Part (99 g) was reduced with lithium aluminum hydride to af- 102.0 (C-4), 107.2 (C-8a), 141.4 (C-4a), 157. 8 (C-1), 157.9 (C-3).
ford 3,5-dibenzyloxybenzyl alcohol (89 g, 97%). A typical reaction condi-
The Naphthalene Type: dH: 6.11 (2, 7-H), 6.35 (4, 5-H); dC: 99.86 (C-2),
The Keto-Tautomer Type: dH: 3.67 (4-H), 5.56 (2-H), 6.15 (5-H), 6.28 (7-
tion for the subsequent step includes the dropwise addition of methanesu- H); dC: 35.0 (C-4), 103.0 (C-7), 104.5 (C-2), 108.6 (C-5), 110.8 (C-8a),
fonyl chloride (0.17 ml) to a dichloromethane solution of the alcohol 143.2 (C-4a), 164.9 (C-6), 165.5 (C-8), 178.0 (C-3), 194.7 (C-1).
(640 mg) in the presence of trimethylamine (0.42 ml) at 4 °C. The mixture
Compound 2 was derivatized to its tetraacetate by treatment with acetic
was stirred for 20 min and poured over iced water and extracted with anhydride in the presence of pyridine. mp 127—129 °C (methanol). Anal.
dichloromethane. The extract was washed with 10% aqueous HCl, saturated Calcd for C18H16O8: C, 60.00; H, 4.48. Found: C, 59.82; H, 4.48. EI-MS
1
aqueous NaHCO3 and brine, followed by drying over Na2SO4. The residue, m/z: 360 (Mϩ), 192 (MϪ4ϫAc). H-NMR (400 MHz, CDCl3) d: 2.33 and
after evaporation of the solvent, was crystallized from dichloromethane– 2.38 (each 3H, s, CH3CO–), 6.99 and 7.51 (each 2H, d, Jϭ2.2, 2.0 Hz, 4-H
hexane to give methanesulfonyl 3,5-dibenzyloxybenzyl alcohol (670 mg, and 5, 7-H).
84%) as colorless needles (mp 80—81 °C). The product (20 g) was dissolved
in chloroform (25 ml), and an aqueous solution (8.5 ml) of sodium cyanide
1,3,8-Trihydroxynaphthalene (T3HN) (4) Scytalone (3) (3.65 g) was
treated with 1 N NaOH (50 ml) at 45 °C under N2 for 3 h. The mixture was
(3.5 g) was added. The mixture was heated under reflux in the presence of poured over iced water, acidified with 0.5 M H3PO4, and extracted with ethyl
benzyl trimethylammonium (813 mg) for 2 h. The cooled mixture was acetate. The extract was washed with water and dried. The residue after
poured into water and extracted with chloroform. The extract was washed evaporation of the solvent was chromatographed on silica gel to give 4
with water and dried over Na2SO4. The residue, after evaporation of the sol- (2.54 g, 76%) as a greenish yellow powder. EI-MS m/z: 176 (Mϩ). HRMS
vent, was subjected to chromatography on silica gel with chloroform– m/z: 176.0476 (Calcd for C10H8O3: 176.0474). The other spectral data were
hexane (4 : 1), followed by crystallization from ethyl acetate–hexane, to give identical with those described in the ref. 3a).
9 (11.4 g, 69%) as pale yellow plates. mp 85—86 °C. Anal. Calcd for
Reduction of T4HN (1) (a) 90 mg of 11 in a methanol (7 ml) solution
C22H19O2N: C, 80.22; H, 5.81; N, 4.25. Found: C, 79.94; H, 5.81; N, 4.34.
of CH3ONa (7.8 mmol, prepared from 180 mg of sodium) was heated at
EI-MS m/z: 329 (Mϩ). 1H-NMR (400 MHz, CDCl3) d: 3.67 (2H, d, 60 °C under N2 for 20 min. After cooling to room temperature, sodium boro-
Jϭ0.7 Hz, –CH2CN), 5.03 (4H, s, –CH2C6H5), 6.15 (3H, br s, 2, 4, 6-H), hydride (80 mg) was added to the mixture. It was further heated at 60 °C
7.2—7.4 (10H, m, 2ϫC6H5).
Methyl(3؅,5؅-dibenzyloxy)phenylacetic Acid (10) 15 grams of 9 was extract was washed with water and dried. The residue after evaporation of
dissolved in dioxane (150 ml) and mixed with 0.5 M Ba(OH)2 (150 ml). The the solvent was chromatographed on consecutive TLCs with benzene–ace-
under N2 for 4 h, poured into 1 N HCl, and extracted with ethyl acetate. The
mixture was heated at 115 °C for 15 h. After cooling, the mixture was acidi- tone (6 : 4) and ether–benzene (5 : 1) to give 3 (18 mg, 20%, the spectral data
fied with 1 N aqueous HCl and extracted with ethyl acetate. The usual work- were identical with those previously described11)) and 6,8-dihydroxy-1-
up and crystallization from ethyl acetate–hexane gave (3Ј,5Ј-dibenzyloxy) methylisochroman-3-one (13) (4.5 mg, 5%). EI-MS m/z: 194 (Mϩ). HRMS
phenylacetic acid (14.5 g, 92%) as colorless needles (mp 106—107 °C). The m/z: 194.0576 (Calcd for C10H10O4: 194.0579). 1H-NMR (400 MHz, ace-
product was converted quantitatively to 10 by treatment with diazomethane. tone-d6) d: 1.53 (3H, d, Jϭ6.8 Hz, 7-H), 3.48 (1H, d, Jϭ9.3 Hz, 2-H), 3.79
mp 63—64 °C (methanol). Anal. Calcd for C23H22O4: C, 76.22; H, 6.12. (1H, d, Jϭ9.3 Hz, 2-H), 5.69 (1H, Jϭ6.8 Hz, 7-H), 6.24 (1H, br s, 3-H), 6.35
Found: C, 76.27; H, 6.20. EI-MS m/z: 362 (Mϩ). 1H-NMR (400 MHz, (1H, dd, Jϭ2.1, 0.8 Hz, 5-H), 8.40 and 8.79 (each 1H, br, –OH).
CDCl3) d: 3.56 (2H, s, –CH2COOCH3), 3.68 (3H, s, OCH3), 5.02 (4H, s,
–CH2C6H5), 6.54 (3H, s, 2, 4, 6-H), 7.3—7.4 (10H, m, 2ϫC6H5).
(b) 38 mg of 2 was treated in essentially the same manner to give scy-
talone (9.5 mg, 25%). The foregoing reduction conditions gave quinonic
Methyl Curvulinate (11) Acetic anhydride (3.6 ml) was added to an compounds in over 60% yield, in which flaviolin (6) was identified by a se-
acetic acid solution (7.5 ml) of 9 (1.1 g), and the mixture was heated at 35 °C ries of spectroscopic studies, as described in ref. 3a).
for 7 min, followed by the dropwise addition of 60% aqueous HClO4
NMR Measurements of T4HN and T3HN under Alkaline Conditions
(6.5 ml). The mixture was further heated at 35 °C for 5 min, poured into Compound 2 (20 mg) was dissolved in a CD3OD (0.6 ml) solution of
water, and extracted with ethyl acetate. The extract was washed with satu- CD3ONa (1.0 mmol) under Ar and transferred into a NMR tube, where a
rated aqueous NaHCO3 and water. The usual work-up and crystallization mixture of CH3OH–CD3OD (1 : 1) was used instead of CD3OD for 2D-NMR
from benzene–hexane yielded methyl curvulinate dibenzylether (1.2 g, 99%) measurements (CH-COSY and HMBC). The tube was sealed under Ar and

196
Vol. 49, No. 2
subjected to NMR measurement. Alternatively, 2 was prepared in situ from
11 or 11a by treatment with CD3ONa (2.0 mmol) in CD3OD (2 ml) at 60 °C
under Ar for 15 min. Part of the mixture (0.6 ml) was used for NMR mea-
surements, as described above. Conversion of 11 to 2, upon stepwise eleva-
tions in temperature, was monitored on a NMR spectrometer by changes in
the H-NMR spectrum of 11 in CD3ONa–CD3OD. H-decoupling was con-
ducted using a JEOL triple resonance system, G50IR2. 1H-decoupled 13C-
NMR spectrum of 4 in CD3ONa–CD3OD gave the following ten signals at
d: 104. 5, 105.7, 113.9, 114.3, 128.1, 128.2, 141.3, 164.6, 165.3, 168.3.
5) Kubo Y., Nakamura H., Kobayashi K., Okuno T., Furusawa I., Plant-
Microbe Interact., 5, 440—445 (1991).
6) Fujii I., Mori Y., Watanabe A., Kubo Y., Tsuji G., Ebizuka Y., Biosci.
Biotechnol. Biochem., 63, 1445—1452 (1999).
7) Anderson J. A., Phytochemistry, 25, 103—106 (1986).
8) Ichinose K., Kiyono J., Ebizuka Y., Sankawa U., Chem. Pharm. Bull.,
41, 2015—2021 (1993).
9) Skory C. D., Chang P. K., Cary J., Linz J. E., Appl. Environ. Micro-
biol., 58, 3527—3537 (1992).
1
2
10) Dimroth P., Walter H., Lynen F., Eur. J. Biochem., 13, 98—110 (1970).
Acknowledgements We are grateful to Prof. H. Noguchi, School of 11) Sankawa U., Shimada H., Sato T., Kinoshita T., Yamasaki K., Chem.
Pharmaceutical Sciences, University of Shizuoka, for his helpful advice in Pharm. Bull., 29, 3536—3542 (1981).
NMR measurement. We also thank Ms. R. Hirata for elemental analysis ser- 12) Wheeler M. H., Exp. Mycol., 6, 171—179 (1982).
vices.
13) Vidal-Cros A., Viviani F., Labesse G., Boccara M., Gaudry M., Eur. J.
Biochem., 219, 985—992 (1994).
Reference and Notes
14) Perpetua N. S., Kubo Y., Yasuda N., Takano Y., Furusawa I., Mol.
Plant Microbe Interact., 9, 323—329 (1996).
15) Tanaka H., Ohne Y., Ogawa N., Tamura T., Agr. Biol. Chem., 27, 48—
55 (1963).
16) Bycroft B. W., Cashyap M. M., Leung T. K., J. Chem. Soc., Chem.
Commun., 1974, 443—444.
1) Preliminary account for this paper was reported: Ichinose K., Ebizuka
Y., Sankawa U., Chem. Pharm. Bull., 37, 2873—2875 (1989).
2) Present address: International Traditional Medicine Research Center,
Toyama International Health Complex, Tomosugi, Toyama 939–8224,
Japan.
3) a) Bell A. A., Stipanovic J. E., Puhalla J. E., Tetrahedron, 32, 1353—
1356 (1976); b) Bell A. A., Puhalia J. E., Tolmsoff W. J., Meola S.,
17) Ichinose K., Sugimori M., Itai A., Ebizuka Y., Sankawa U., Tetrahe-
dron Lett., 31, 5905—5908 (1990).
Can. J. Microbiol., 22, 702—711 (1976); c) Bell A. A., Wheeler W. 18) Ichinose K., Sugimori M., Itai A., Ebizuka Y., Sankawa U., Chem.
H., Annu. Rev. Phytopathol., 24, 411—451 (1986).
Lett., 1991, 219—222.
4) Hopwood D. A., Chem. Rev., 97, 2465—2497 (1997).