Gene Disruption and Biochemical Characterization of Verruculogen Synthase of Aspergillus fumigatus

Article abstract of DOI:10.1002/cbic.201000562

Three in one: Verruculogen synthase FtmF is a key enzyme to diversify the fumitremorgin biosynthetic pathway in Aspergillus fumigatus: FtmF catalyzes not only endoperoxide bond formation of verruculogen but also the deprenylation and oxidation of fumitremorgin B to yield 12α,13α-dihydroxyfumitremorgin C and 13-oxoverruculogen, respectively.

Full text of DOI:10.1002/cbic.201000562

DOI: 10.1002/cbic.201000562
Gene Disruption and Biochemical Characterization of Verruculogen
Synthase of Aspergillus fumigatus
Naoki Kato,[a] Hirokazu Suzuki,[a, b] Hiroshi Takagi,[a] Masakazu Uramoto,[a] Shunji Takahashi,[a] and
Hiroyuki Osada*[a]
The completion of more than ten fungal genomes has re-
vealed the presence of an enormous number of genes that
encode proteins typically involved in secondary-metabolite
biosynthesis, such as polyketide synthase and nonribosomal
peptide synthetase.[1] These genes, which are responsible for
forming the backbone structures of natural products, are often
clustered with the genes encoding so-called tailoring enzymes,
which are responsible for modifications of the metabolites,
such as prenylation, methylation, and hydroxylation.[2] Their
gene organization allows us to predict biosynthetic products
that arise from the corresponding gene clusters. Based on such
predictions, many biosynthetic gene clusters and their metabo-
lites have been identified by various approaches, such as het-
erologous expression, gene knockout and complementation
analyses, and silent-gene activation.[3] On the other hand, pre-
diction of the function of each gene in the clusters is not
always successful because only the types of protein functions
can be annotated by a homology search; this is not sufficient
for the correct prediction of substrates or products of un-
known enzymes. Therefore, functional analyses of every gene
in biosynthetic gene clusters are needed to discover the en-
zymes involved in important processes of natural product bio-
synthesis.
protein belonging to the a-ketoglutarate (a-KG)-dependent di-
oxygenases, is a candidate for such missing enzymes. Very re-
cently, Steffan et al. showed that FtmF catalyzes the conversion
of fumitremorgin B (1) to verruculogen (2) by using recombi-
nant protein (Scheme 1).[8] This biosynthetic step is unique in
that FtmF catalyzes the formation of an endoperoxide bond
between two isoprene moieties. However, no genetic evidence
for the involvement of ftmF in the biosynthesis of 2 in A. fumi-
gatus has been provided. The FtmF reaction itself remains to
be fully understood due to the paucity of biochemical data.
Here, we perform a knockout experiment and a kinetic analysis
to clarify the function of FtmF as an endoperoxidation catalyst.
Furthermore, metabolite profiling with a gene disruption
mutant and the identification of FtmF reaction products reveal
that FtmF plays an important role in giving rise to the structur-
al diversity of fumitremorgins.
We first compared the metabolite profiles of the ftmF dis-
ruption mutant to that of the wild-type strain. The DftmF
strain was previously generated from the strain TAFK1.39 (Da-
kuA::ptrA) derived from
a fumitremorgin-producer stain,
BM939, by replacing the entire coding region with the hy-
gromycin B-resistance gene cassette.[7] The transformants were
cultivated in order to analyze their metabolite profiles by LC/
ESI-MS. The disruption of ftmF stopped the production of 2
but had little or no effect on the production of 1 and its bio-
synthetic intermediates (Figure 1); this demonstrates that ftmF
is involved in the biosynthesis of 2 in A. fumigatus.
For biochemical characterization, the recombinant FtmF was
expressed in Escherichia coli as a His6-tagged fusion protein
and purified to homogeneity by metal ion affinity chromatog-
raphy (Figure S2). Size-exclusion column chromatography sug-
The ftm cluster, which is responsible for fumitremorgin pro-
duction in Aspergillus fumigatus (Figure S1A in the Supporting
Information), is one of the secondary metabolite biosynthetic
gene clusters whose products have been identified based on
genomic information.[4,5] Biochemical and genetic studies have
demonstrated the functions of most of the ftm genes: an NRPS
gene, ftmA;[5] two prenyltransferase genes, ftmB and ftmH (also
known as ftmPT1 and ftmPT2);[4,6] and three cytochrome P450
genes, ftmC, ftmE, and ftmG.[7] In spite of these extensive stud-
ies, there are still some missing steps in the biosynthetic path-
way to fumitremorgins that are predicted to involve oxygenas-
es. Our previous analysis of the ftm cluster revealed that three
cytochrome P450s are responsible for key steps in the fumitre-
morgin biosynthetic pathway (Figure S1B),[7] pointing out that
the only remaining oxygenase gene, ftmF, which encodes a
[a] Dr. N. Kato, Dr. H. Suzuki, H. Takagi, Dr. M. Uramoto, Dr. S. Takahashi,
Dr. H. Osada
Chemical Biology Department, RIKEN Advanced Science Institute
Wako, Saitama 351-0198 (Japan)
Fax: (+81)48-462-4669
Figure 1. Metabolite profiles of the wild-type and the DftmF strains derived
from A. fumigatus BM939. The fungal strains were cultivated at 288C for
72 h. The culture extracts were analyzed by LC/ESI-MS. UV detection was car-
ried out at 220 nm. The production was analyzed independently in two dis-
ruption mutants of ftmF. Retention times of authentic standards of 1, 2, and
4 are denoted by Arabic numerals. The peak indicated with an asterisk was
16 mass unit larger than 1, but its chemical structure was not determined.
The biosynthetic intermediates of 1 are indicated by closed circles.
E-mail: hisyo@riken.jp
[b] Dr. H. Suzuki
Present address: Organization of Advanced Science and Technology
Kobe University, Hyogo 657-8501 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201000562.
ChemBioChem 2011, 12, 711 – 714
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
711

Scheme 1. The reactions catalyzed by FtmF. Molecular oxygen attacks the substrate radical 1a that is generated by reduction of the reactive FeIV =O species
in the active center of FtmF to FeIII-OH, resulting in the formation of the peroxide radical 1b, followed by cyclization to yield 2. Ascorbate is used for the re-
duction of radical species to yield the FtmF products, such as 2 and 3a, as well as the regeneration of FeIII to FeII.
gested that FtmF was likely to be homodimer (77 kDa); this is
in agreement with previous reports.[8,9] The production of 2
was clearly detected when FtmF was incubated with 1 in the
presence of a-KG, FeII, and ascorbate (0.55 mmolminÀ1 per mg
of protein, Figure 2). The structure of 2 was confirmed by mass
spectrometry and NMR analysis. a-KG and FeII were required
for the formation of 2 by FtmF, whereas ascorbate was not es-
sential but was involved in the FtmF reaction (Figure S3A); this
is consistent with a previous report.[8] FtmF catalysis had an
optimum pH of 6.0, and more than 90% of enzymatic activity
was retained after heat treatment at 378C for 10 min. The ki-
netic parameters for FtmF were measured by monitoring the
formation of 2 by HPLC. The FtmF reaction apparently fol-
lowed Michaelis–Menten kinetics. The Km values of 1 and a-KG
were 21.3Æ4.8 and 3.2Æ0.5 mm, respectively, with a kcat value
of 11.5Æ0.4 minÀ1.
The endoperoxide bond of 2 has been proposed to be
formed through radical formation at C-21 of 1, followed by in-
corporation of molecular oxygen (Scheme 1);[8] this is similar to
the proposed mechanism for peroxide bond formation by cy-
clooxygenase (COX), which is responsible for the biosynthesis
of prostaglandin G2,[11] but not a typical hydroxylation mecha-
nism of a-KG dioxygenases.[12] The possible mechanism that
involves sequential monooxygenation has been excluded be-
cause the product 2, which contains two atoms of either 18O
Figure 2. In vitro reaction of FtmF. A) HPLC chromatograms of the FtmF re-
action (bottom) and the authentic standard of 2 that was prepared from the
culture broth of A. fumigatus BM939 (top). UV detection was carried out at
300 nm. The peak indicated with an asterisk was 16 mass units larger than
1, but its chemical structure was not determined. B) Time course of the
FtmF reaction. Relative abundance of the substrate 1 and the FtmF products
or 16O but not one of each, was formed by FtmF under 18O2-en-
riched conditions.[8] We also performed the FtmF reaction in a
reaction chamber where the air was replaced by the isotopical-
ly stable 18O2 and analyzed the reaction products by LC/ESI-MS.
Ion peaks four mass units larger than those of the authentic
2–4 was plotted (amount of 1 at t=0 was defined as 100).
712
www.chembiochem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2011, 12, 711 – 714

standard of 2 were detected (Figure S4), thus supporting the
proposed reaction mechanism of FtmF.
tional catalysis, which might lead to the development of bio-
synthetic tools to modify natural products for their structural
diversity.
Under the optimized assay condition, several peaks in addi-
tion to 2 were detected in the reaction mixture of FtmF with 1
(Figure 2A). To determine the structures of the reaction prod-
ucts, we performed a large-scale reaction (Figure S3B), and the
isolated compounds were characterized by mass spectrometry
and NMR analysis. Two compounds corresponding to the
peaks 3 and 4 were identified as 12a,13a-dihydroxyfumitre-
morgin C and 13-oxoverruculogen, respectively. Compound 3
is a biosynthetic product that is converted from fumitremor-
gin C by FtmG[7] as well as a substrate of FtmH to form 1 (Fig-
ure S1B).[6] We speculate that this deprenylation could proceed
via a hemiaminal structure (3a) that might be formed by the
reaction of radical intermediate 1a with H2O. The oxidized
product 4 was isolated from the culture broth of A. fumi-
gatus.[12] Metabolite profiling of the ftmF disruption mutant
also showed ftmF-dependent production of 4 in A. fumigatus
(Figure 1). These results indicate that this compound is pro-
duced during the normal biosynthetic process in A. fumigatus,
rather than as a by-product formed under artificial conditions.
No significant peak including 4 was detected when FtmF was
incubated with 2 as a substrate (Figure S3C); this indicates that
4 is formed through oxidation at C-13 to yield 4a, followed by
endoperoxide bond formation (Scheme 1). However, the reac-
tion mechanism of this oxidation remains unclear. Besides 4,
other fumitremorgin analogues with a carbonyl group at C-13,
such as 13-oxofumitremorgin B (4a) and cyclotryprostatin D,
were reported to be isolated from the culture extracts of A. fu-
migatus.[12,13] Our data provide evidence that FtmF is also in-
volved in this missing step of the fumitremorgin pathway in
A. fumigatus. Peaks with m/z of 478 were detected at retention
times of 15.2 min in the FtmF reaction products (Figure 2A)
and 21.2 min in the culture extract of the wild type (Figure 1).
The corresponding compounds should be monooxygenated
products of 1 (m/z 462 [M+HÀH2O]+), though their structures
could not be identified in this study.
Experimental Section
Fungal strains and culture conditions: The ftmF disruption
mutant of A. fumigatus and its parental strain, TAFK1.39 (wild-type),
were described previously.[7] For the analysis of fumitremorgin pro-
duction, freshly harvested spore suspensions of the strains were in-
oculated in Aspergillus glucose minimum medium[15] supplemented
with yeast extract (0.5%, w/v). The cultures were cultivated at 288C
for 72 h and cleared by filtration. The culture filtrates were extract-
ed with ethyl acetate. The dried extracts were dissolved in metha-
nol and analyzed by HPLC and LC/ESI-MS. The conditions for HPLC
and LC/ESI-MS were described previously.[7]
For the preparation of 1 and 2, A. fumigatus BM939 was cultivated
at 288C for 120 h in fermentation medium.[7] The fungal culture
was cleared by filtration and extracted by ethyl acetate. From the
dried extract, compounds were isolated by normal-phase chroma-
tography on silica 60N (Kanto chemicals, Tokyo, Japan) followed by
preparative HPLC. The structures of isolated compounds were con-
firmed by mass spectrometry and NMR analysis. Their NMR spectra
were identical to the data previously reported.[7,16]
Vector construction, heterologous expression, and enzyme pu-
rification: The ftmF gene was amplified by PCR using chromoso-
mal DNA from A. fumigatus BM939 as a template. The primer pair
used was 5’-AGG ATC CAT GCA TCA TCA CCA TCA CCA TAT GAC
CGT CGA CTC CAA GCC-3’ (BamHI) and 5’-ACT CGA GTT AAG CCG
GCG AAT CAG CAG-3’ (XhoI). The amplified DNA fragment was
cloned in pCR4Blunt-TOPO (Invitrogen) and verified by sequencing.
The ORF of ftmF was then cloned into the BamHI-XhoI site of
pCold II (Takara, Otsu, Japan), resulting in pCold II–ftmF. This plas-
mid contained ftmF under the cold shock-inducible cspA promoter.
E. coli BL21(DE3) carrying pCold II-ftmF was cultured at 378C for
6 h in LB medium supplemented with ampicillin (50 mgmLÀ1). Fol-
lowing the addition of IPTG (0.1m) and incubation at 158C for
30 min, the culture was further cultured at 158C for 24 h. The cells
were harvested, suspended in buffer A [sodium phosphate (50 mm,
pH 8.0), NaCl (0.5m), glycerol (20%, v/v), and 2-mercaptoethanol
(0.1%, v/v)] containing imidazole (10 mm) and lysozyme
(2 mgmLÀ1), and incubated on ice for 30 min. From the suspen-
sion, cell lysates were obtained by sonication, followed by centrifu-
gation. The clear lysates were applied onto a Talon-Co2+ column
(Clontech) equilibrated with buffer A containing imidazole (10 mm).
The column was washed with the same buffer. Recombinant FtmF
was eluted with buffer A containing imidazole (250 mm). The
buffer was exchanged for buffer B [Tris-HCl (50 mm, pH 7.5), glycer-
ol (20%, v/v), and 2-mercaptoethanol (0.1%, v/v)], and the purified
enzyme was concentrated.
The time course of the FtmF reaction showed that the for-
mation of 2 was approximately ten times faster than that of 3
(0.048 mmolminÀ1 per mg of protein; Figure 2B). The formation
of 4 was much less than that of 3. This result indicates that the
endoperoxide bond formation of 2 is the main function of
FtmF, and that it is also capable of catalyzing deprenylation at
N-1 and oxidation at C-13 of 1 to yield 3 and 4 as minor prod-
ucts, respectively.
Our knockout experiment and kinetic analysis clearly dem-
onstrate that a-KG dioxygenase FtmF is responsible for the for-
mation of the endoperoxide bond of 2 in A. fumigatus. Further-
more, characterization of the FtmF reaction products reveals
that FtmF also catalyzes the deprenylation at N-1 and the oxi-
dation at C-13 of 1. Several bioactive natural products contain-
ing endoperoxide bonds, such as the antimalarial drug artemi-
sinin,[14] have been reported so far. Despite growing interest in
the chemistry of endoperoxide formation, only two enzymes,
FtmF and COX, have been reported to catalyze endoperoxide
bond formation to date. A crystallographic study of FtmF is es-
sential to further understand the molecular basis of multifunc-
Molecular mass analysis: The native molecular mass was estimat-
ed on a gel filtration column (Superdex 200, GE Healthcare) by fast
protein liquid chromatography with isocratic elution at a flow rate
of 0.2 mLminÀ1 in buffer B containing NaCl (0.15m). Aldolase
(158 kDa), conalbumin (75 kDa), ovalbumin (43 kDa), carbonic an-
hydrase (29 kDa), ribonuclease A (13.7 kDa), and aprotinin (6.5 kDa)
were used for the molecular mass standards.
Enzyme assay: The standard reaction mixture (400 mL) consisted
of the recombinant FtmF (2 mg), 1 (50 mm), Fe(NH4)3(SO4)2 (1 mm),
ascorbate (1 mm), a-KG (1 mm), and MES-NaOH buffer (50 mm,
pH 6.0) containing glycerol (20%, v/v) and 2-mercaptoethanol
ChemBioChem 2011, 12, 711 – 714
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
713

(0.1%, v/v). For kinetic analysis, 0.67 mg of the enzyme was used.
The substrate concentrations were varied from 10 to 66.7 mm for 1
and 1 to 200 mm for a-KG. After preincubation of the reaction mix-
tures at 308C for 5 min, the reactions were initiated by the addi-
tion of enzyme and were allowed to proceed for 3 min. The reac-
tions were terminated by the addition of EDTA (final concentration
of 0.1m). Reaction products were extracted with ethyl acetate and
analyzed by HPLC and LC/ESI-MS. The conditions for HPLC and LC/
ESI-MS were described previously.[7] The amounts of the substrate
1, and the reaction products 2 and 3 were calculated with stan-
dard curves, which were obtained from their maximum UV wave-
lengths. Compounds 1–3 were subjected to HPLC at a concentra-
tion range of 0.1 to 12.5 nmol for their calibration. The amounts of
4 were estimated roughly with the standard curve of 1. The
enzyme-specific activity (mmol of product formed per min per mg
of enzyme) for each product was calculated by time-dependent
product formation. The kinetic constants were calculated by non-
linear regression fit to the Michaelis–Menten equation. Kinetic con-
stants that were calculated from Lineweaver–Burk, Eadie–Hofstee,
and Hanes–Woolf plots were not significantly different from those
derived from nonlinear regression.
20), 123.2 (d, C-16), 118.9 (s, C-15), 117.3 (d, C-22), 111.9 (d, C-17),
108.8 (s, C-14), 95.2 (d, C-19), 86.8 (d, C-21), 81.9 (s, C-27), 81.4 (s,
C-12), 60.2 (d, C-6), 55.7 (q, 18-OCH3), 53.5 (t, C-26), 48.2 (d, C-3),
45.6 (t, C-9), 28.5 (t, C-7), 26.6 (q, C-28), 25.7 (q, C-24), 23.8 (q, C-
29), 23.4 (t, C-8), 19.0 (q, C-25); (+)-ESI-MS (m/z): 532 [M+Na]+, 510
[M+H]+. The NMR spectra were in good agreement with the re-
ported data,[10] except for the assignment of the chemical shifts.
We reassigned the chemical shifts by 2D NMR.
Enzyme assay in the presence of stable isotope 18O2: Molecular
oxygen 18O2 (99%) was purchased from Taiyo Nippon Sanso (Tokyo,
Japan). The reaction mixture (1 mL) consisted of the same compo-
nents as those for the standard assay. Before starting the FtmF re-
action, the reaction mixture in the sealed chamber was degassed
and flushed with 18O2. The reaction was carried out at 308C for
20 min and analyzed by LC/ESI-MS.
Acknowledgements
We are grateful to Dr. T. Shimizu, Dr. A. Kawasaki, and Dr. Y.
Nakai for valuable advice. We also thank Dr. T. Saito and Dr. H.
Konno (Akita Konno Co. Ltd.) for providing the fungal broth. This
work was supported in part by a Grant-in-Aid for Creative Scien-
tific Research from the Ministry of Education, Culture, Sports, Sci-
ence, and Technology of Japan, and by funding from the Special
Postdoctoral Researchers Program of RIKEN (to N.K. and H.S.).
Characterization of the reaction products of FtmF: For structural
elucidation of the reaction products,
a large-scale reaction
(100 mL) was carried out by using the purified FtmF (86 mg) and 1
(84 mm) in Tris-HCl buffer (50 mm, pH 7.0) containing the same sup-
plements as for the standard assay at 308C for 24 h. The reaction
products 2–4 were isolated by preparative HPLC, and their struc-
tures were determined from the following spectroscopic parame-
ters.
1
Keywords: Aspergillus fumigatus · biosynthesis · dioxygenase ·
endoperoxide bond · natural products
Verruculogen (2): H NMR (500 MHz, CDCl3): d=7.88 (d, J=8.7 Hz,
1H), 6.81 (dd, J=8.7, 2.3 Hz, 1H), 6.62 (d, J=8.3 Hz, 1H), 6.58 (d,
J=2.3 Hz, 1H), 6.03 (d, J=10.1 Hz, 1H), 5.64 (brs, 1H), 5.02 (brdt,
J=8.3, 1.4 Hz, 1H), 4.74 (d, J=2.3 Hz, 1H), 4.47 (dd, J=9.9, 7.3 Hz,
1H), 3.99 (s, 1H), 3.83 (s, 3H), 3.63 (brdd, J=8.9, 4.1 Hz, 2H), 2.48
(m, 1H), 1.98 (d, J=1.4 Hz, 3H), 1.85–2.14 (m, 4H), 1.72 (d, J=
1.4 Hz, 3H), 1.70 (s, 3H), 1.67 (dd, J=13.3, 10.1 Hz, 1H), 1.01 (s,
3H); (+)-ESI-MS (m/z): 534 [M+Na]+, 494 [M+HÀH2O]+. The NMR
spectra were in good agreement with those of the compounds iso-
lated from the culture broth.
[1] N. P. Keller, G. Turner, J. W. Bennett, Nat. Rev. Microbiol. 2005, 3, 937–
947.
[2] N. Khaldi, F. T. Seifuddin, G. Turner, D. Haft, W. C. Nierman, K. H. Wolfe,
N. D. Fedorova, Fungal Genet. Biol. 2010, 47, 736–741.
[3] A. A. Brakhage, V. Schroeckh, Fungal Genet. Biol. 2011, 48, 15–22; K.
Scherlach, C. Hertweck, Org. Biomol. Chem. 2009, 7, 1753–1760.
[4] A. Grundmann, S. M. Li, Microbiology 2005, 151, 2199–2207.
[5] S. Maiya, A. Grundmann, S. M. Li, G. Turner, ChemBioChem 2006, 7,
1062–1069.
[6] A. Grundmann, T. Kuznetsova, S. S. Afiyatullov, S. M. Li, ChemBioChem
2008, 9, 2059–2063.
[7] N. Kato, H. Suzuki, H. Takagi, Y. Asami, H. Kakeya, M. Uramoto, T. Usui, S.
Takahashi, Y. Sugimoto, H. Osada, ChemBioChem 2009, 10, 920–928.
[8] N. Steffan, A. Grundmann, S. Afiyatullov, H. Ruan, S. M. Li, Org. Biomol.
Chem. 2009, 7, 4082–4087.
[9] E. Eichhorn, J. R. van der Ploeg, M. A. Kertesz, T. Leisinger, J. Biol. Chem.
1997, 272, 23031–23036.
[10] L. J. Marnett, Curr. Opin. Chem. Biol. 2000, 4, 545–552.
[11] C. Krebs, D. Galonic Fujimori, C. T. Walsh, J. M. Bollinger, Jr., Acc. Chem.
Res. 2007, 40, 484–492.
[12] F. Wang, Y. C. Fang, T. J. Zhu, M. Zhang, A. Lin, Q. Q. Gu, W. M. Zhu, Tet-
rahedron 2008, 64, 7986–7991.
[13] C. B. Cui, H. Kakeya, H. Osada, Tetrahedron 1997, 53, 59–72.
[14] N. J. White, Science 2008, 320, 330–334.
12a,13a-Dihydroxyfumitremorgin C (3): 1H NMR (500 MHz, CDCl3):
d=7.78 (d, J=8.7 Hz, 1H), 7.71 (brs, 1H), 6.82 (d, J=1.8 Hz, 1H),
6.79 (dd, J=8.7, 2.3 Hz, 1H), 5.85 (dd, J=9.6, 1.4 Hz, 1H), 5.72 (dd,
J=2.8, 1.4 Hz, 1H), 4.78 (dt, J=9.6, 1.4, 1.4 Hz, 1H), 4.66 (d, J=
2.8 Hz, 1H), 4.41 (dd, J=10.1, 6.9 Hz, 1H), 4.16 (brs, 1H), 3.82 (s,
3H), 3.63 (m, 2H), 2.47 (m, 1H), 2.10 (m, 2H), 1.98 (d, J=1.4 Hz,
3H), 1.95 (m, 1H), 1.64 (d, J=1.4 Hz, 3H); (+)-ESI-MS (m/z): 434
[M+Na]+, 394 [M+HÀH2O]+; (À)-ESI-MS (m/z): 410 [MÀH]À, 392
[MÀHÀH2O]À. The NMR spectra were identical to the previously re-
ported data.[7]
1
13-Oxoverruculogen (4): H NMR (500 MHz, CDCl3): d=8.20 (d, J=
8.7 Hz, 1H; H-16), 6.95 (dd, J=8.7, 2.2 Hz, 1H; H-17), 6.65 (d, J=
8.0 Hz, 1H; H-21), 6.59 (d, J=2.2 Hz, 1H; H-19), 6.30 (dd, J=9.9,
1.0 Hz, 1H; H-3), 5.00 (brdt, J=8.0, 1.3, 1.3 Hz, 1H; H-22), 4.86 (t,
J=8.3, 8.3 Hz, 1H; H-6), 3.83 (s, 3H; 18-OCH3), 3.60 (m, 2H; H-9),
2.40 (m, 1H; H-7b), 2.21 (m, 1H; H-7a), 2.01 (d, J=1.3 Hz, 3H; H-
25), 1.95 (m, 3H; H-8, H-26a), 1.82 (dd, J=14.0, 1.0 Hz, 1H; H-26b),
1.76 (d, J=1.3 Hz, 3H; H-24), 1.68 (s, 3H; H-28), 1.00 (s, 3H; H-29);
13C NMR (125 MHz, CDCl3): d=181.6 (s, C-13), 173.0 (s, C-5), 165.1
(s, C-11), 157.9 (s, C-18), 147.8 (s, C-2), 145.5 (s, C-23), 137.3 (s, C-
[15] N. Kato, W. Brooks, A. M. Calvo, Eukaryotic Cell 2003, 2, 1178–1186.
[16] J. Fayos, D. Lokensgard, J. Clardy, R. J. Cole, J. W. Kirksey, J. Am. Chem.
Soc. 1974, 96, 6785–6787.
Received: September 19, 2010
Published online on February 18, 2011
714
www.chembiochem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2011, 12, 711 – 714