DOI: 10.1002/cbic.200800787
Identification of Cytochrome P450s Required for
Fumitremorgin Biosynthesis in Aspergillus fumigatus
Naoki Kato,[a] Hirokazu Suzuki,[a, d] Hiroshi Takagi,[a] Yukihiro Asami,[a, e] Hideaki Kakeya,[a, f]
Masakazu Uramoto,[a] Takeo Usui,[b] Shunji Takahashi,[a] Yoshikazu Sugimoto,[c] and
Hiroyuki Osada*[a]
Introduction
Breast cancer resistance protein (BCRP) is a member of the
multidrug transporters of the ATP-binding cassette family,
which can actively extrude a wide range of structurally diverse
drugs, toxins, endogenous compounds, and their metabolites
across the plasma membranes of cells.[1, 2] Although these ATP-
dependent efflux pumps were once thought to be of relevance
only to multidrug resistance in cancer cells, it is now clear that
they have a pronounced role in the pharmacokinetics of a
broad range of drugs and toxins. It is noteworthy that recent
findings have revealed intriguing roles for BCRP in stem
cells.[2, 3] Specific inhibitors are therefore required for further un-
derstanding of the pharmacological and physiological roles of
this interesting transporter in normal and malignant stem cells,
as well as of clinical applications of BCRP inhibition in cancer
chemotherapy. Two BCRP inhibitors—GF120918 and fumitre-
morgin C (6)—have been well characterized. GF120918 is a
synthetic product originally developed as a P-glycoprotein in-
hibitor.[4, 5] In contrast, compound 6, a diketopiperazine myco-
toxin produced by Aspergillus fumigatus,[6] is capable of com-
pletely reversing mitoxantrone, doxorubicin, and topotecan re-
sistance in BCRP-overexpressing cells but does not reverse re-
sistance to cells that overexpress other multidrug transporters
such as P-glycoprotein or multidrug-resistant protein 1.[7,8] Sev-
eral synthetic analogues of 6 have been investigated and one
such, Ko143, showed more potent inhibitory effects as well as
lower in vivo toxicity.[9, 10] This indicates that 6 may serve as a
lead compound for more potent and specific BCRP inhibitors.
Elucidation of the fumitremorgin biosynthetic pathway pro-
vides a strategy for new drug design.
The A. fumigatus genome harbors more than 20 biosynthetic
gene clusters for secondary metabolites.[11] Their gene organi-
zation allows us to predict biosynthetic products that arise
from the corresponding gene clusters. It has been suggested
that some of the gene clusters are involved in the biosynthesis
of known fungal metabolites.[12–14] The ftm gene cluster has
also been investigated as the most probable candidate for the
biosynthesis of 6 and its related compounds.[15–17] It apparently
consists of nine genes (Figure 1A). A genetic study has indicat-
ed that the dimodular nonribosomal peptide synthetase
(NRPS) gene ftmA encodes brevianamide F synthetase.[17] Func-
Fumitremorgin C, a diketopiperazine mycotoxin produced by
Aspergillus fumigatus, is a potent and specific inhibitor of
breast cancer resistance protein (BCRP). Elucidation of the
ACHTUNGTRENNUNGfumitremorgin C biosynthetic pathway provides a strategy for
new drug design. A structure–activity relationship study based
on metabolites related to the ftm gene cluster revealed that
the process most crucial for inhibitory activity against BCRP
was cyclization to form fumitremorgin C. To determine the
gene involved in the cyclization reaction, targeted gene inacti-
vation was performed with candidate genes in the ftm cluster.
Analysis of the gene disruptants allowed us to identify ftmE,
one of the cytochrome P450 genes in the cluster, as the gene
responsible for the key step in fumitremorgin biosynthesis.
ACHTUNGTRENNUNG dditionally, we demonstrated that the other two cytochrome
P450 genes, ftmC and ftmG, were involved in hydroxylation of
the indole ring and successive hydroxylation of fumitremor-
gin C, respectively.
[a] Dr. N. Kato,+ Dr. H. Suzuki,+ H. Takagi, Dr. Y. Asami, Dr. H. Kakeya,
Dr. M. Uramoto, Dr. S. Takahashi, Dr. H. Osada
Chemical Biology Department, Advanced Science Institute, RIKEN
Wako, Saitama 351-0198 (Japan)
Fax: (+81)48-462-4669
E-mail : hisyo@riken.jp
[b] Dr. T. Usui
Graduate School of Life and Environmental Sciences, University of Tsukuba
Ibaraki 305-8572 (Japan)
[c] Dr. Y. Sugimoto
Division of Chemotherapy, Faculty of Pharmacy, Keio University
Tokyo 105-8512 (Japan)
[d] Dr. H. Suzuki+
Present address: Organization of Advanced Science and Technology
Kobe University
Hyogo 657-8501 (Japan)
[e] Dr. Y. Asami
Present address: Graduate School of Pharmaceutical Sciences
The University of Tokyo
Tokyo 113-0033 (Japan)
[f] Dr. H. Kakeya
Present address: Division of Bioinformatics and Chemical Genomics
Graduate School of Pharmaceutical Sciences, Kyoto University
Kyoto 606-8501 (Japan)
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.200800787.
920  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2009, 10, 920 – 928

tional analyses of FtmB and FtmH (also termed FtmPT1 and
FtmPT2, respectively) have been performed to characterize
their enzymatic activities.[15,16] These results have suggested
that this gene cluster directs the biosynthesis of fumitremor-
gin B (8), with ftmA, ftmB, and ftmH involved in the first,
second, and last steps, respectively, in the biosynthetic path-
way to 8. However, the functions of the other ftm genes
remain to be elucidated. Lack of fumitremorgin production in
the genome reference strain Af293[17] makes a full understand-
ing of the cluster difficult, so in exploring the fumitremorgin
pathway we utilized the strain BM939, which is a high produc-
er of 6 and its related compounds.[18]
In the work reported here we carried out a structure–activity
relationship (SAR) study, revealing that the most crucial event
for exertion of inhibitory activity against BCRP was the CN
bond formation in the synthesis of 6. To identify the gene
ACHTUNGTRENNUNGresponsible for the key step to form 6, targeted gene inactiva-
tion for candidate genes in the ftm cluster was performed.
Analysis of the knockout mutants allowed us to identify the cy-
tochrome P450 gene ftmE as involved in the CN bond forma-
tion. In addition, we demonstrated the role of the other two
cytochrome P450 genes—ftmC and ftmG—in the fumitremor-
gin pathway.
Results and Discussion
Structure–activity relationship study based on metabolites
related to the ftm cluster
To examine the biological activities of 6 and its related com-
pounds, we isolated metabolites associated with the ftm clus-
ter from BM939, a fumitremorgin-producing strain of A. fumi-
gatus. Eight diketopiperazine compounds—brevianamide F (1),
tryprostatin B (2), demethoxyfumitremorgin C (3), desmethyl-
tryprostatin A (4), tryprostatin A (5), fumitremorgin C (6),
12a,13a-dihydroxyfumitremorgin C (7), and fumitremorgin B
(8)—were prepared, and their structures were determined by
mass spectrometry and NMR analysis. Compound 4, a des-
methyl analogue of 5, is a new compound. Disruption of ftmA
in the BM939 strain caused a deficiency in the production of
1–8 (Figure 1B), indicating that these metabolites are products
of the ftm cluster
These metabolites share a diketopiperazine scaffold but are
structurally diverse and thereby useful for SAR studies of their
bioactivities. In fact, evaluation of 1–8 with regard to BCRP
ACHTUNGTRENNUNGinhibitory activity (Figure 2A and B) revealed that 6 was the
most potent inhibitor out of all of the derivatives tested. The
SAR study also indicates that the following three moieties
were important for the inhibitory activity of 6 against BCRP
(Figure 2C). 1) The most crucial moiety involved in the activity
of 6 is the covalent bond between C-3 and N-4, because com-
pounds 3 and 6–8 all showed detectable activities in assays in
vivo and in vitro, whereas the activities of 1, 2, 4, and 5 were
negligible. Although compound 5 has been reported as a
BCRP inhibitor,[19] its activity was much lower than that of 6
under the conditions used in this study. 2) Dihydroxylation at
C-12 and C-13 of 6 impaired inhibitory activity at the cellular
level. Reversal effects of 1–6, but not of 7 or 8, on drug resist-
Figure 1. The production of metabolites associated with the ftm cluster in
A. fumigatus. A) Gene organization of the ftm cluster. The three cytochro-
me P450 genes are indicated in black. The genes shown in gray—ftmA/
ftmPS, ftmB/ftmPT1, and ftmH/ftmPT2—encode a dimodular NRPS[17] and pre-
nyltransferases.[15, 16] B) HPLC chromatograms of culture extracts of the wild-
type and the ftmA strains derived from A. fumigatus BM939. UV detection
was carried out at 220 nm. Retention times of authentic standards of fumi-
tremorgins are denoted by Arabic numerals. MS analysis confirmed that the
peak at a retention time of 14.5 min in the chromatograms of extracts of
the ftmA strains contained no compound 4.
ChemBioChem 2009, 10, 920 – 928  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.chembiochem.org 921
Cytochrome P450 Genes in the ftm Gene Cluster

ance in BCRP-overexpressing K562 cells was in good agree-
ment with their inhibitory activities against BCRP-dependent
ATPase activities. A possible explanation for the weaker effects
of 7 and 8 in the in vivo assay is a change in membrane
ACHTUNGTRENNUNGpermeability due to further modifications. 3) Compound 6
showed clear inhibitory activities even at submicro-
molar concentrations, whereas its demethoxy form 3
did not (data not shown), which suggests that the
methoxy group at C-18 is important, in agreement
with previous suggestions.[9]
Although fumitremorgins are harmful tremorgenic
mycotoxins produced by A. fumigatus and related
fungi,[6] some of the biosynthetic intermediates be-
sides the BCRP inhibitor 6 have been shown to have
interesting biological and pharmacological activi-
ties.[18,20] Recently, Jain et al. reported that 5, which
has inhibitory effects on the cell cycle[18] and micro-
tubule assembly,[21] and its synthetic derivatives
showed insignificant bioactivities.[22] Consistently
with this, such inhibitory effects were not detectable
in the compounds isolated in this study.
Identification of the key enzymes for fumitremor-
gin biosynthesis
The SAR study of fumitremorgins 1–8 revealed the
important moieties involved in exertion of the BCRP
inhibitory activity of 6. From this information we
tried to identify the genes involved in the formation
of such important moieties. None of the enzymes
catalyzing the cyclization to form 6, the subsequent
hydroxylation at C-12 and C-13 of 6, or the hydroxyl-
ation at C-6 of the indole ring had been previously
identified. To identify the genes responsible for
these reactions, we first cloned the ftm cluster from
the strain BM939. A 27 kb DNA fragment that covered the ftm
cluster of strain BM939 was sequenced, revealing that the clus-
ter is extremely similar to that of Af293 and consists of nine
genes (see Table 1 for features of the ftm gene products).
There were six uncharacterized genes in the ftm cluster. FtmC,
Figure 2. Evaluation of BCRP inhibitory effects of fumitremorgins. A) Inhibition of drug
efflux in K562/BCRP cells by fumitremorgins 1–8. IC50 values of growth inhibition of
K562/BCRP cells by SN-38 in the presence of fumitremorgins (3 mm) were determined.
IC50 values in K562/BCRP and K562 cells (parental cells) without fumitremorgins were de-
fined as 0 and 100% inhibition, respectively. B) Inhibition of BCRP-dependent ATPase ac-
tivity by fumitremorgins 1–8. The vanadate-sensitive ATPase activities in the presence of
fumitremorgins (50 mm) were measured in vitro with use of BCRP membranes (BD Bio-
sciences). C) SAR of fumitremorgins. Of the moieties shown in gray, the covalent bond
between C-3 and N-4 and the methoxy group at C-18 are important for the inhibitory ac-
tivity of 6 against BCRP, whereas dihydroxylation at C-12 and C-13 of 6 affects the activity
at the cellular level.
Table 1. Features of the ftm gene products of A. fumigatus BM939.
Protein Size exon Function[a] Relatives[b] (identity/similarity [%]) Accession
bp/aa number
FtmA 6636/2211 1–6636 dimodular NRPS nonribosomal peptide synthetase XyNRPSA ABF29402
from Xylaria sp. BCC 1067 (37/55)
FtmC 1955/559 1–969, 1033–1154, 1227–1525, cytochrome P450 isotrichodermin C-15 hydroxylase TRI11 from O13317
1597–1698, 1768–1955 Fusarium sporotrichioides (31/46)
FtmD 1114/342 1–528, 614–1114 methyltransferase cercosporin toxin biosynthesis protein CTB3 ABC79591
from Cercospora nicotianae (31/51)
FtmB 1464/464 1–1262, 1332–1464 prenyltransferase dimethylallyltryptophan synthase DmaW from AAP81210
Claviceps purpurea (34/56)
FtmE 1581/526 1–1581 cytochrome P450 cytochrome P450 ELN2 from Coprinopsis cinerea BAA33717
ACHTUNGTRENNUNG(26/44)
FtmF 876/291 1–876 a-KG dioxygenase fumonisin C-5 hydroxylase Fum3p/FUM9 from AAG27131
Gibberella moniliformis (29/48)
FtmG 1813/504 1–207, 273–389, 451–550, cytochrome P450 GA14-synthase P450-1 from G. fujikuroi (37/56) CAA75565
604–672, 723–1313, 1383–1813
FtmH 1349/427 1–1181, 1247–1349 prenyltransferase tryptophan dimethylallyltransferase FgaPT2 from AAX08549
A. fumigatus (37/56)
FtmI 2043/680 1–2043 protein–protein ankyrin 1 isoform 2 from Homo sapiens (35/53) NP_065210
interaction
[a] Functions of FtmD, FtmF, and FtmI were predicted on the basis of sequence similarities to known proteins. a-KG: a-ketoglutarate. [b] The listed homol-
ogous proteins exclude putative proteins derived from genomic projects.
922 www.chembiochem.org  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2009, 10, 920 – 928
H. Osada et al.

FtmE, and FtmG show similarity to cytochrome P450s of fila-
mentous fungi, while FtmF has similarity to proteins that
belong to the a-ketoglutarate dioxygenase family.[23] These
ACHTUNGTRENNUNGenzymes could have roles in the cyclization as well as in the
hydroxylations. Besides the oxygenases, FtmD is predicted to
function as an O-methyltransferase and is thus implicated in
the methylation of the new intermediate 4 to give 5. The ftmI
gene encodes an ankyrin-repeat protein.[24]
To assign the roles of the oxygenase genes in fumitremorgin
biosynthesis, we generated gene disruptants by replacing the
entire coding region of each gene with the hygromycin B-
resistance gene cassette (Dftm ::hph ; Figure 3). The akuA strain
(TAFK1.39), derived from BM939, was used as a host strain for
the knockout experiments, and correct disruption events oc-
curred in almost all hygromycin-resistant transformants (data
not shown). Two to four transformants of each ftm disruption
were cultivated for analysis of fumitremorgin production. The
metabolite profiles of the disruptants—ftmC , ftmE , ftmF ,
and ftmG—were determined by HPLC and LC/ESI-MS (see
Table S1 in the Supporting Information for productivity of
ACHTUNGTRENNUNGfumitremorgins in the disruptants).
The disruption of ftmC led to substantial accumulation of 2
and its cyclization product 3 (Figure 4A). Compound 4 (the
product hydroxylated at C-6 of the indole ring of 2) and its
downstream methoxy-group-containing metabolites 5–8 were
not detected in the culture extracts of the ftmC strain. The
ftmE disruptants produced 2 and 5 but not their cyclization
products 3 and 6 (Figure 4B). The disruption of ftmG resulted
in the loss of production of 7 and 8 (Figure 4C). The produc-
tion of 6 was observed in the ftmG strain, indicating that hy-
droxylation of the indole ring and cyclization proceeded nor-
mally in this strain. These results clearly suggest the roles of
the three cytochrome P450 genes in the fumitremorgin path-
way: hydroxylation at C-6 of the indole ring, CN bond forma-
tion to form 6, and the subsequent dihydroxylation are medi-
ated by ftmC, ftmE, and ftmG, respectively. On the other hand,
the disruption of ftmF had no significant effect on the produc-
tion of 1–8, suggesting that ftmF should not be involved in
their biosynthesis (data not shown).
On the basis of the phenotypes of the ftm disruptants, the
functions of the three cytochrome P450 genes were demon-
strated by use of a yeast expression system. The cytochro-
me P450 genes ftmC, ftmE, and ftmG were expressed in Sac-
charomyces cerevisiae with a P450 reduction partner gene of
A. fumigatus, AFUA_2g07940. Microsomes that were prepared
from ftmC-expressing yeast catalyzed the hydroxylation of 2 to
Figure 3. Construction of the ftm disruptants. DNA fragments (5.9kb) containing 1 kb upstream and 1 kb downstream regions of the ftm genes and the
ACHTUNGTRENNUNGhygromycin B-resistance cassette (hph) were used for the transformation of the wild-type strain (WT, TAFK1.39) and as probes for Southern hybridization.
A) ftmA disruption: total DNA (10 mg) isolated from the hygromycin B-resistant transformants was digested with a) MluI, or b) ApaI. The WT strain shows
a) 16.1, and b) 10.3 and 5.4 kb bands, whereas the ftmA mutant shows a) 9.7 and 3.5, and b) 12.8 kb bands. B) ftmC disruption: total DNA was digested with
a) NdeI, or b) Aor51HI. WT shows a) 8.8, and b) 6.0 and 2.9 kb bands, whereas the mutant shows a) 5.9 and 4.7, and b) 10.7 kb bands. C) ftmE disruption: total
DNA was digested with a) NheI+XbaI, or b) KpnI. WT shows a) 5.3, and b) 3.5 and 3.4 kb bands, whereas the mutant shows a) 4.2 and 3.3, and b) 9.1 kb
bands. D) ftmF disruption: total DNA was digested with a) NdeI, or b) NruI. WT shows a) 12.9, and b) 5.2 and 3.7 kb bands, whereas the mutant shows a) 10.3
and 5.5, and b) 11.8 kb bands. E) ftmG disruption: total DNA was digested with a) SacII, or b) SmaI. WT shows a) 11.1, and b) 4.3 and 1.9 kb bands, whereas the
mutant shows a) 7.3 and 5.5, and b) 8.2 kb bands.
ChemBioChem 2009, 10, 920 – 928  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.chembiochem.org 923
Cytochrome P450 Genes in the ftm Gene Cluster

yield 4 in the presence of NADPH (Figure 5A), even though its
expression level was not high enough for the CO spectrum to
be detectable. The functions of FtmE and FtmG were evaluat-
ed by bioconversion experiments: ftmE-expressing yeast cells
converted 5 into 6 effectively (13 nmh1), whereas they also
converted 2 into the shunt product 3 (6.5 nmh1; Figure 5B).
Presumably these cyclizations proceeded through hydroxyl-
ation at C-18 or N-10 by FtmE, followed by dehydration to
form the CN bond. To the best of our knowledge, this is the
first fungal cytochrome P450 that catalyzes CN bond forma-
tion. Other than this, there is only one bacterial cytochro-
me P450, StaN, that catalyzes CN bond formation between
aglycon and deoxysugar moieties during staurosporine biosyn-
thesis in Streptomyces sp. TP-A0274.[25] The conversion of 6 into
7 by ftmG-expressing yeast was observed, though the conver-
sion rate was very low (4.6 nmday1; Figure 5C).
Proposed biosynthetic pathway for fumitremorgins
Previous studies have already pointed out that three genes in
the ftm cluster—ftmA, ftmB, and ftmH—are involved in the
first, second, and last steps, respectively, in the biosynthetic
pathway of 8.[15–17] The first committed step of the fumitremor-
gin pathway is the formation of 1—diketopiperazine formation
from two amino acids, l-tryptophan and l-proline. This was
further supported by the lack of production of 1–8 that was
observed in the ftmA disruptants derived from BM939 (Fig-
ure 1B). Heterologous expression of ftmA conferred the ability
to produce 1 both to S. cerevisiae (data not shown) and to
A. nidulans,[17] so 1 was obviously the biosynthetic product
ACHTUNGTRENNUNGattributable to ftmA and was the precursor of 2–8. The sub-
ACHTUNGTRENNUNGsequent step is the prenylation of 1 to form 2 by FtmB/
FtmPT1.[16] The other prenyltransferase, FtmH/FtmPT2, catalyz-
es the prenylation of the indole ring at N-1 of 7 to yield 8.[15]
Figure 4. The metabolite profiles of the ftm disruptants derived from A. fumigatus BM939. HPLC chromatograms of culture extracts of the knockout mutants
of A) ftmC, B) ftmE, and C) ftmG. The fungal strains were cultivated at 28 8C for 48 h. Fumitremorgins 1–8 in the culture extracts were determined by HPLC
and LC/ESI-MS with reference to authentic standards. The production was analyzed independently in two to four clones of each strain. UV detection was car-
ried out at 220 nm. D) Proposed biosynthetic pathway of fumitremorgins 3–7.
924 www.chembiochem.org  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2009, 10, 920 – 928
H. Osada et al.

Our results elucidated the missing link in the fumitremorgin
pathway, which is composed of four processes (Figure 4D): the
hydroxylation of the indole ring of 2 at C-6 by FtmC, followed
by methylation to form 5, the CN bond formation for the syn-
thesis of 6 by FtmE, and the subsequent hydroxylation of 6 at
C-12 and C-13 by FtmG. There are three genes—ftmD, ftmF,
and ftmI—in the cluster that remain to be characterized. Be-
cause the predicted function of FtmD is that of a methyltrans-
ferase, FtmD is a plausible candidate for the enzyme that cata-
lyzes the methylation of 4 to form 5.
There are several fumitremorgin-related compounds that
could not be accounted for by the ftm gene functions. One
such compound is verruculogen, which contains a unique epi-
dioxy (C-O-O-C) bridge in its structure.[6] To date, there is no
report of enzymes that catalyze epidioxy formation except for
prostaglandin endoperoxide H synthase.[26] Because the func-
tions of three out of four oxygenase genes in the cluster—
ftmC, ftmE and ftmG—were determined, the last uncharacter-
ized oxygenase gene, ftmF, might be a candidate for this inter-
esting reaction. FtmF-dependent peroxidation is now under in-
vestigation. The disruption of ftmI had no significant effect on
the production of 1–8 (data not shown), indicating that ftmI
was unlikely to be involved in the biosynthesis of 1–8.
Figure 5. Reconstitution of the cytochrome P450-mediated reactions with a yeast expression system. A) HPLC chromatogram of reaction products of FtmC.
Microsomes that were prepared from the yeast cells expressing ftmC and AFUA_2g07940 were incubated with 2 (50 mm) in the presence of NADPH (1 mm) at
30 8C for 60 min. UV detection was carried out at 300 nm. B) HPLC chromatograms of culture extracts of the yeast cells expressing ftmE and AFUA_2g07940.
The cells were incubated with substrates 5 (upper panels) and 2 (lower panels ; each 2.5 mm) at 30 8C for two days. UV detection was carried out at 280 and
300 nm for reaction products of 2 and 5, respectively. C) HPLC chromatograms of culture extracts of yeast cells expressing ftmG and AFUA_2g07940. The cells
were incubated with substrate 6 (2.5 mm) at 30 8C for two days. UV detection was carried out at 300 nm.
ChemBioChem 2009, 10, 920 – 928  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.chembiochem.org 925
Cytochrome P450 Genes in the ftm Gene Cluster

Conclusions
Our SAR study on metabolites associated with the ftm cluster
demonstrated that fumitremorgin C (6) was the most potent
inhibitor against BCRP of all of the metabolites that were
tested. A crucial moiety for exertion of the inhibitory activity of
6 was the covalent bond between C-3 and N-4. Methoxylation
of the indole ring at C-6 and the dihydroxylation at C-12 and
C-13 also modulated inhibitory activity. Targeted gene inacti-
ACHTUNGTRENNUNGvation with a fumitremorgin producer strain, BM939, revealed
that the three cytochrome P450 genes—ftmC, ftmE, and
ftmG—are involved in these biosynthetic processes. We con-
firmed their enzymatic activities with a yeast expression
system. In particular, the FtmE-mediated oxidative ring-closure
step is noteworthy. To the best of our knowledge, this enzyme
is the first fungal cytochrome P450 that catalyzes CN bond
formation. This study has elucidated the missing links in the fu-
mitremorgin pathway, which are also crucial processes for ex-
ertion of the inhibitory activity of 6 against BCRP, not only pro-
viding insights into mycotoxin biosynthesis but also opening
the way to improved biosynthesis of intermediates that have
interesting pharmacological activities.
Experimental Section
Microbial strains and plasmids : A. fumigatus BM939 was isolated
previously.[18] The cosmid AN26, which contains the hygromycin B-
resistant cassette,[27] was obtained from the Fungal Genetics Stock
Center. E. coli strains TOP10 and DH5a and plasmids pCR2.1-TOPO,
pCR4Blunt-TOPO, pDONR P4-P1R/P2R-P3/221, and pDEST R4-R3 (In-
vitrogen) were used for DNA manipulation. S. cerevisiae YPH500
and pESC-URA (Stratagene) were used for heterologous expression
of the ftm genes.
Preparation of fumitremorgins : A. fumigatus BM939 was cultivat-
ed at 28 8C for 3–5 days in complete medium [malt extract (2%),
Bacto peptone (1%), glucose (2%)] . The fungal culture was cleared
by filtration and extracted with ethyl acetate. From the dried ex-
tract, fumitremorgins were isolated by normal-phase chromatogra-
phy on silica 60N (Kanto chemicals) followed by preparative HPLC.
Their structures were determined from the following spectroscopic
parameters.
Brevianamide F (1): 1H NMR (500 MHz, CDCl3): d=8.24 (br s, 1H),
7.57 (d, J=7.8 Hz, 1H), 7.38 (d, J=8.1 Hz, 1H), 7.21 (td, J=7.3,
0.9 Hz, 1H), 7.12 (td, J=7.3, 0.9 Hz, 1H), 7.09 (d, J=1.8 Hz, 1H),
5.73 (br s, 1H), 4.36 (dd, J=11.0, 2.8 Hz, 1H), 4.05 (t, J=7.3 Hz, 1H),
3.74 (ddd, J=15.1, 3.7, 0.9 Hz, 1H), 3.60 (m, 2H), 2.95 (dd, J=15.1,
11.0 Hz, 1H), 2.30 (m, 1H), 1.99 (m, 2H), 1.89 ppm (m, 1H); ESI-MS:
m/z : 284.1 [M+H]+ . The NMR spectra were identical to the report-
ed data.[16]
Tryprostatin B (2): 1H NMR (500 MHz, CDCl3): d=7.94 (br s, 1H), 7.46
(d, J=7.8 Hz, 1H), 7.29 (d, J=8.3 Hz, 1H), 7.14 (ddd, J=7.6, 7.1,
1.3 Hz, 1H), 7.08 (ddd, J=10.6, 7.8, 1.0 Hz, 1H), 5.59 (br s, 1H), 5.30
(t, J=7.0 Hz, 1H), 4.35 (brdd, J=11.5, 2.8 Hz, 1H), 4.04 (t, J=
7.8 Hz, 1H), 3.64 (m, 2H), 3.57 (ddd, J=11.8, 9.4, 3.2 Hz, 1H), 3.49
(dd, J=17.6, 7.8 Hz, 1H), 3.44 (dd, J=16.5, 6.9 Hz, 1H), 2.93 (dd,
J=15.1, 11.5 Hz, 1H), 2.31 (m, 1H), 2.05–2.00 (m, 2H), 1.95–1.85 (m,
1H), 1.77 (s, 3H), 1.74 ppm (s, 3H); ESI-MS: m/z : 352.1 [M+H]+ . The
NMR spectra were identical to the reported data.[28]
Demethoxyfumitremorgin C (3): 1H NMR (500 MHz, CDCl3): d=7.79
(br s, 1H), 7.56 (d, J=7.3 Hz, 1H), 7.33 (d, J=8.1 Hz, 1H), 7.17 (br t,
J=6.3 Hz, 1H), 7.13 (br t, J=6.9 Hz, 1H), 6.01 (d, J=9.6 Hz, 1H),
4.90 (brd, J=9.1 Hz, 1H), 4.17 (dd, J=11.7, 5.0 Hz, 1H), 4.10 (br t,
J=8.2 Hz, 1H), 3.63 (m, 2H), 3.55 (dd, J=16.0, 5.0 Hz, 1H), 3.11
(dd, J=15.8, 11.5 Hz, 1H), 2.40 (m, 1H), 2.23 (m, 1H), 2.05 (m, 1H),
2.00 (s, 3H), 1.94 (m, 1H), 1.63 ppm (s, 3H); ESI-MS: m/z : 350.3
[M+H]+ . The NMR spectra were identical to the reported data.[28]
Desmethyltryprostatin A (4): Pale yellow powder; [a]21D =25.5 (c
0.25, in methanol) ; 1H NMR (500 MHz, [D6]DMSO): d=10.30 (s, 1H),
8.73 (s, 1H), 7.19 (d, J=8.6 Hz, 1H), 7.03 (s, 1H), 6.62 (d, J=2.3 Hz,
1H), 6.42 (dd, J=2.3, 8.6 Hz, 1H), 5.27 (t, J=7.0 Hz, 1H), 4.19 (t, J=
5.1 Hz, 1H), 3.98 (brdd, J=7.3, 9.3 Hz, 1H), 3.45 (dd, J=7.0,
15.5 Hz, 1H), 3.39 (m, 1H), 3.28 (dd, J=7.0, 15.5 Hz, 1H), 3.17 (d,
J=4.5 Hz, 1H), 3.14 (dd, J=5.1, 14.5 Hz, 1H), 2.90 (dd, J=6.5,
14.5 Hz, 1H), 1.89 (m, 1H), 1.69 (s, 3H), 1.68 (s, 3H), 1.61 (m, 1H),
1.42 (m, 1H), 1.15 ppm (m, 1H); 13C NMR (125 MHz, [D6]DMSO): d=
168.3 (s), 165.4 (s), 152.4 (s), 136.4 (s), 134.7 (s), 131.7 (s), 121.9 (d),
121.3 (s), 118.4 (d), 108.4 (d), 103.8 (s), 96.1 (d), 58.4 (d), 55.2 (d),
44.5 (t), 27.5 (t), 26.2 (t), 25.5 (q), 24.8 (t), 21.6 (t), 17.7 ppm (q); UV/
Vis : lmax=222, 273, 299 nm; HR-FAB-MS: m/z : calcd for C21H26N3O3:
368.1974 [M+H]+ ; found: 368.1980; ESI-MS: m/z : 368.3 [M+H]+ .
Tryprostatin A (5): 1H NMR (500 MHz, CDCl3): d=7.79 (br s, 1H), 7.32
(d, J=8.7 Hz, 1H), 6.81 (d, J=2.3 Hz, 1H), 6.74 (dd, J=8.7, 2.3 Hz,
1H), 5.61 (br s, 1H), 5.28 (t, J=7.3 Hz, 1H), 4.31 (brdd, J=11.2,
2.7 Hz, 1H), 4.04 (t, J=7.8 Hz, 1H), 3.81 (s, 3H), 3.63 (ddd, J=10.2,
8.0, 3.7 Hz, 1H), 3.62 (dd, J=15.1, 4.1 Hz, 1H), 3.57 (ddd, J=11.9,
8.9, 2.8 Hz, 1H), 3.43 (dd, J=17.4, 6.8 Hz, 1H), 3.39 (dd, J=16.5,
7.3 Hz, 1H), 2.89 (dd, J=15.1 11.5 Hz, 1H), 2.31 (m, 1H), 2.05–1.98
(m, 2H), 1.89 (m, 1H), 1.76 (d, J=1.0 Hz, 3H), 1.73 ppm (s, 3H); ESI-
MS: m/z : 382.3 [M+H]+ . The NMR spectra were identical to the re-
ported data.[28]
Fumitremorgin C (6): 1H NMR (500 MHz, CDCl3): d=7.65 (br s, 1H),
7.42 (d, J=8.3 Hz, 1H), 6.84 (d, J=2.3 Hz, 1H), 6.79 (dd, J=8.7,
2.3 Hz, 1H), 5.96 (d, J=9.6 Hz, 1H), 4.89 (dt, J=9.6, 1.4 Hz, 1H),
4.17 (dd, J=11.9, 4.6 Hz, 1H), 4.09 (t, J=7.8 Hz, 1H), 3.82 (s, 3H),
3.62 (m, 2H), 3.50 (dd, J=16.0, 5.0 Hz, 1H), 3.08 (ddd, J=15.8, 11.5,
0.9 Hz, 1H), 2.38 (m, 1H), 2.22 (m, 1H), 2.02 (m, 1H), 1.98 (d, J=
0.7 Hz, 3H), 1.94 (m, 1H), 1.63 ppm (d, J=1.4 Hz, 3H); ESI-MS: m/z :
380.2 [M+H]+ . The NMR spectra were identical to the reported
data.[28]
12a,13a-Dihydroxyfumitremorgin C (7): 1H NMR (500 MHz, CDCl3):
d=7.78 (d, J=8.7 Hz, 1H), 7.64 (br s, 1H), 6.83 (d, J=2.3 Hz, 1H),
6.78 (dd, J=8.7, 2.3 Hz, 1H), 5.85 (dd, J=9.4, 0.9 Hz, 1H), 5.73 (dd,
J=2.8, 0.9 Hz, 1H), 4.78 (dt, J=9.6, 1.4 Hz, 1H), 4.64 (d, J=2.8 Hz,
1H), 4.41 (dd, J=10.1, 6.9 Hz, 1H), 4.09 (br s, 1H), 3.82 (s, 3H), 3.62
(m, 2H), 2.47 (m, 1H), 2.08 (m, 1H), 2.02 (m, 1H), 1.99 (d, J=1.4 Hz,
3H), 1.95 (m, 1H), 1.65 (d, J=0.9 Hz, 3H); ESI-MS: m/z : 394.2
[M+HH2O]+ . The NMR spectra were identical to the reported
data.[29]
Fumitremorgin B (8): 1H NMR (500 MHz, CDCl3): d=7.83 (d, J=
8.7 Hz, 1H), 6.78 (dd, J=8.7, 2.3 Hz, 1H), 6.67 (d, J=1.8 Hz, 1H),
5.97 (d, J=10.1 Hz, 1H), 5.75 (s, 1H), 5.02 (br t, J=6.9 Hz, 1H), 4.68
(brd, J=10.1 Hz, 1H), 4.52 (br s, 2H), 4.43 (dd, J=9.9, 7.3 Hz, 1H),
3.82 (s, 3H), 3.62 (dd, J=8.9, 4.6 Hz, 2H), 2.46 (m, 1H), 2.20–1.90
(m, 3H), 1.97 (d, J=1.4 Hz, 3H), 1.83 (s, 3H), 1.68 (d, J=1.0 Hz, 3H),
1.61 (d, J=1.4 Hz, 3H); ESI-MS: m/z : 462.1 [M+HH2O]+ . The NMR
spectra were identical to the reported data.[30]
BCRP inhibitory assay : The BCRP inhibitory activities of fumitre-
morgins 1–8 were assessed by growth inhibition of K562 cells that
926 www.chembiochem.org  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2009, 10, 920 – 928
H. Osada et al.

overexpressed the BCRP gene (K562/BCRP) by the anticancer drug
SN-38, as described previously.[31] Briefly, K562/BCRP cells were
grown in RPMI 1640 medium that was supplemented with fetal
bovine serum (7%, v/v) at 37 8C in CO2 (5%, v/v). The sensitivity of
the K562/BCRP cells to SN-38 in the presence of fumitremorgins
(3 mm) was evaluated by cell growth inhibition after incubation at
37 8C for 4 days. Cell numbers were determined with a Coulter
counter. The IC50 values (drug dose that caused 50% inhibition of
cell growth) were determined from the growth inhibition curves.
The inhibitory effects of fumitremorgins 1–8 on BCRP activity were
also evaluated in vitro by measuring BCRP-dependent ATPase
ACHTUNGTRENNUNGactivity, as described previously,[32] with minor modifications. BCRP
membranes (BD Biosciences) were incubated at 37 8C in medium
(95 mL) consisting of Tris-MES (50 mm, pH 6.8), EGTA (2 mm), DTT
(2 mm), KCl (50 mm), sodium azide (5 mm), and fumitremorgins
(50 mm). The ATPase reaction was started by the addition of MgATP
(100 mm, 5 mL). To measure BCRP-independent ATPase activity, an
identical reaction mixture that contained sodium orthovanadate
(400 mm) was assayed in parallel. After incubation for 30 min, re-
ACHTUNGTRENNUNGactions were terminated by addition of perchloric acid (0.6m,
100 mL). The amount of inorganic phosphate was determined as
described previously.[33]
Cloning of the ftm cluster of A. fumigatus BM939 : An AflII site
was introduced into the cloning site of a cosmid vector (Super-
Cos1, Stratagene). The resulting vector was used for construction
of a genomic library of A. fumigatus BM939 with AflII-digested
chromosomal DNA of the strain. On screening of the library, a
27 kb cosmid clone that covered the ftm genes was isolated.
Disruption of the ftm genes : We first prepared the akuA-disrupted
strain derived from A. fumigatus BM939. The akuA gene encodes
the Ku70 component that causes low efficiency of homologous
ACHTUNGTRENNUNGrecombination in filamentous fungi.[34,35] For construction of the
akuA knockout plasmids, 1 kb DNA fragments upstream of the
start codon and downstream of the stop codon of akuA were
ACHTUNGTRENNUNGamplified by PCR with use of chromosomal DNA of A. fumigatus
BM939 as template. The primer pairs akuA-UF ACHTUNGTRENNUNG(1023)/akuA-UR-
ACHTUNGTRENNUNG(16) and akuA-DFACHTUNGTRENNUNG(2269)/akuA-DRACHTUNGTRENNUNG(3265) were used for amplifica-
tion of the upstream and downstream regions, respectively. The
pyrithiamine-resistant gene ptrA[36] was used as a selection marker
for the akuA knockout. These DNA fragments were combined in
the original orientation in pDEST by use of the MultiSite Gateway
System (Invitrogen) in the following order: the upstream region,
ptrA, followed by the downstream region. From this plasmid, a
DNA fragment (4.0 kb) was excised by KpnI digestion and used for
transformation of A. fumigatus BM939. Pyrithiamine-resistant trans-
formants (DakuA ::ptrA) that resulted from double-crossover be-
tween the disrupted akuA sequence and the intact chromosomal
akuA sequence were isolated. Correct disruption was checked by
Southern hybridization (data not shown). The resulting akuA
strain TAFK1.39 was used as a recipient strain for further transfor-
mations.
Knockout mutants of the ftm genes were prepared from TAFK1.39,
in a procedure similar to that described for the akuA disruption.
The 1 kb DNA fragments upstream and downstream of the ftm
genes were amplified by PCR with use of chromosomal DNA of
BM939 as template. The primer pairs ftm-UF and -UR and ftm-DF
and -DR were used for amplification of the upstream and down-
stream regions, respectively. The hygromycin B-resistant cassette
(hph) was used as a selection marker. These DNA fragments were
combined in the original orientation in pDEST in the following
order: the upstream regions, hph, followed by the downstream re-
gions. From these plasmids, 5.9 kb DNA fragments were excised by
restriction enzyme digestion and used for fungal transformation.
The restriction enzymes that were used are indicated in Table S2.
Hygromycin B-resistant transformants (Dftm ::hph) were verified by
genomic Southern analysis to contain the ftm gene replacements
(Figure 3). Note that the parent akuA strain TAFK1.39 is described
as the “wild-type” strain in this study. All of the DNA fragments am-
plified by PCR were verified by sequencing. The oligonucleotides
that were used for PCR are summarized in Table S2.
Determination of fumitremorgins produced by the ftm disrup-
tants : Freshly harvested spore suspensions of an A. fumigatus
strain were inoculated in fermentation medium [K2HPO4 (0.5%),
MgSO4·7H2O (0.05%), soybean meal (2%), glucose (3%), soluble
starch (2%), pH 6.5] . The culture was cultivated at 28 8C for 48 h
and cleared by filtration. The culture filtrate was extracted with
ethyl acetate. The dried extracts were dissolved in methanol and
analyzed by HPLC and LC/ESI-MS.
HPLC analysis was carried out with a Waters 600 HPLC system with
a photodiode array detector (2996 PDA detector). The HPLC condi-
tions were as follows: column, Senshu Pak Docosil-B 3 m (4.6
250 mm); flow rate, 1.0 mLmin1; solvent A, water containing
formic acid (0.05%, v/v) ; solvent B, acetonitrile. After injection of
the sample into a column equilibrated with solvent B (25%), the
column was developed with a linear gradient from 25% to 65%
over 20 min, followed by isocratic elution of solvent B (65%) for
20 min. LC/ESI-MS analysis was carried out with a Waters Alliance
HPLC system fitted with a mass spectrometer (Q-TRAP, Applied Bio-
systems). The HPLC conditions were as follows: column, Senshu
Pak Docosil-B 3 m (2.0250 mm, Senshu Scientific) ; flow rate,
0.2 mLmin1. After injection of the sample into a column equili-
brated with solvent B (10%), the column was developed with a
linear gradient from 10% to 100% solvent B over 90 min. Mass
spectra were collected in an ESI-positive mode.
Construction of plasmids for heterologous expression of the ftm
genes : The ORFs of AFUA_2g07940 and ftmE were amplified by
PCR with chromosomal DNA of A. fumigatus BM939 as template.
The ORFs of ftmC and ftmG were amplified by two-step RT-PCR
with total RNA extracted from A. fumigatus BM939 as template. All
DNA fragments amplified by PCR were cloned into pCR4Blunt-
TOPO and verified by sequencing. The AFUA_2g07940 ORF in
pCR4Blunt-TOPO was excised by SalI–XhoI digestion and cloned
into the SalI-XhoI site of pESC-URA, resulting in pEUR07940. The
ORFs of ftmC, ftmE, and ftmG were cloned in the NotI–SpeI site of
pEUR079490, resulting in pEUR07940-ftmC, -ftmE, and -ftmG, re-
spectively. These plasmids contained AFUA_2g07940 and the ftm
gene under the GAL1 and GAL10 promoters, respectively. The oli-
gonucleotides that were used for PCR are summarized in Table S3.
In vitro assay of FtmC : S. cerevisiae YPH500 containing
pEUR07940-ftmC was cultivated at 30 8C for three days in SGI
medium [yeast nitrogen base (0.7%), galactose (2%), casamino
acids (0.1%), with l-tryptophan and l-histidine (20 mgL1), l-leu-
cine (30 mgL1), and adenine (200 mgL1)] . From the harvested
cells, microsomes were prepared as described previously.[37] The CO
spectrum was undetectable. The reaction mixture (500 mL) consist-
ed of Tris·HCl (50 mm, pH 7.5), glycerol (20%, v/v), 2-mercaptoetha-
nol (15 mm), fumitremorgin substrate (50 mm), NADPH (1 mm), and
microsomes. After the reaction mixtures had been incubated at
30 8C for 60 min, the reactions were terminated by addition of HCl
(a final concentration of 0.1m). Reaction products were extracted
with ethyl acetate and analyzed by HPLC and LC/ESI-MS.
ChemBioChem 2009, 10, 920 – 928  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.chembiochem.org 927
Cytochrome P450 Genes in the ftm Gene Cluster

Bioconversion assay of FtmE and FtmG : S. cerevisiae YPH500 car-
rying pEUR07940-ftmE or -ftmG was cultivated at 30 8C for 1 day in
SGI medium. After fumitremorgin substrates had been added to
the cultures (final concentrations of 2.5 mm), the cultures were fur-
ther incubated for two days. The compounds in the broths were
extracted with ethyl acetate and analyzed by HPLC and LC/ESI-MS.
The following conditions were used for HPLC analysis of the reac-
tion products of in vitro and bioconversion assays: column, Senshu
Pak Docosil-B (4.6250 mm); flow rate, 1.0 mLmin1. After injec-
tion of the sample into a column equilibrated with 20% solvent B,
the column was initially developed isocratically for 3 min. The
column was successively developed with a linear gradient 20% to
100% over 15 min, isocratic elution for 1 min, a linear gradient
100% to 20% over 1 min, followed by isocratic elution of solvent B
(20%) over 10 min.
Accession numbers : The nucleotide sequence reported in this
paper has been deposited to the GenBank/DDBJ/EMBL database
under accession number AB436628.
Acknowledgements
We are grateful to Y. Koyama (Noda Institute for Scientific Re-
search) for valuable discussions. We also thank S. Simizu, H. Ichi-
miya, and S. Kazami for evaluation of the bioactivities of the
compounds, and T. Saito for advice and support. This work was
supported in part by a Grant-in-Aid for Creative Scientific Re-
search from the Ministry of Education, Culture, Sports, Science,
and Technology of Japan, and by funding from the Special Post-
doctoral Researchers Program of RIKEN (to N.K. and H.S.).
Keywords: Aspergillus fumigatus · biosynthesis ·
cytochrome P450 · fumitremorgins · natural products
[1] L. A. Doyle, D. D. Ross, Oncogene 2003, 22, 7340–7358.
[2] P. Krishnamurthy, J. D. Schuetz, Annu. Rev. Pharmacol. Toxicol. 2006, 46,
381–410.
[3] S. Zhou, J. D. Schuetz, K. D. Bunting, A. M. Colapietro, J. Sampath, J. J.
Morris, I. Lagutina, G. C. Grosveld, M. Osawa, H. Nakauchi, B. P. Sorrenti-
no, Nat. Med. 2001, 7, 1028–1034.
[4] M. de Bruin, K. Miyake, T. Litman, R. Robey, S. E. Bates, Cancer Lett. 1999,
146, 117–126.
[5] F. Hyafil, C. Vergely, P. Du Vignaud, T. Grand-Perret, Cancer Res. 1993, 53,
4595–4602.
[6] R. J. Cole, M. A. Schweikert in Handbook of Secondary Fungal Metabo-
lites, Vol. 1 (Eds. : R. J. Cole, M. A. Schweikert, B. B. Jarvis), Academic
Press, San Diego, 2003, pp. 222–232.
[7] S. K. Rabindran, H. He, M. Singh, E. Brown, K. I. Collins, T. Annable, L. M.
Greenberger, Cancer Res. 1998, 58, 5850–5858.
[8] S. K. Rabindran, D. D. Ross, L. A. Doyle, W. Yang, L. M. Greenberger,
Cancer Res. 2000, 60, 47–50.
[9] J. D. Allen, A. van Loevezijn, J. M. Lakhai, M. van der Valk, O. van Tellin-
gen, G. Reid, J. H. Schellens, G. J. Koomen, A. H. Schinkel, Mol. Cancer
Ther. 2002, 1, 417–425.
[10] A. van Loevezijn, J. D. Allen, A. H. Schinkel, G. J. Koomen, Bioorg. Med.
Chem. Lett. 2001, 11, 29–32.
[11] W. C. Nierman, A. Pain, M. J. Anderson, J. R. Wortman, H. S. Kim, J.
Arroyo, M. Berriman, K. Abe, D. B. Archer, C. Bermejo, J. Bennett, P.
Bowyer, D. Chen, M. Collins, R. Coulsen, R. Davies, P. S. Dyer, M. Farman,
N. Fedorova, N. Fedorova, T. V. Feldblyum, R. Fischer, N. Fosker, A.
Fraser, J. L. Garcia, M. J. Garcia, A. Goble, G. H. Goldman, K. Gomi, S. Grif-
fith-Jones, R. Gwilliam, B. Haas, H. Haas, D. Harris, H. Horiuchi, J. Huang,
S. Humphray, J. Jimenez, N. Keller, H. Khouri, K. Kitamoto, T. Kobayashi,
S. Konzack, R. Kulkarni, T. Kumagai, A. Lafon, J. P. Latge, W. Li, A. Lord, C.
Lu, W. H. Majoros, G. S. May, B. L. Miller, Y. Mohamoud, M. Molina, M.
Monod, I. Mouyna, S. Mulligan, L. Murphy, S. O’Neil, I. Paulsen, M. A. Pe-
nalva, M. Pertea, C. Price, B. L. Pritchard, M. A. Quail, E. Rabbinowitsch,
N. Rawlins, M. A. Rajandream, U. Reichard, H. Renauld, G. D. Robson, S.
Rodriguez de Cordoba, J. M. Rodriguez-Pena, C. M. Ronning, S. Rutter,
S. L. Salzberg, M. Sanchez, J. C. Sanchez-Ferrero, D. Saunders, K. Seeger,
R. Squares, S. Squares, M. Takeuchi, F. Tekaia, G. Turner, C. R. Vazquez
de Aldana, J. Weidman, O. White, J. Woodward, J. H. Yu, C. Fraser, J. E.
Galagan, K. Asai, M. Machida, N. Hall, B. Barrell, D. W. Denning, Nature
2005, 438, 1151–1156.
[12] D. M. Gardiner, B. J. Howlett, FEMS Microbiol. Lett. 2005, 248, 241–248.
[13] S. Maiya, A. Grundmann, X. Li, S. M. Li, G. Turner, ChemBioChem 2007, 8,
1736–1743.
[14] I. A. Unsold, S. M. Li, Microbiology 2005, 151, 1499–1505.
[15] A. Grundmann, T. Kuznetsova, S. S. Afiyatullov, S. M. Li, ChemBioChem
2008, 9, 2059–2063.
[16] A. Grundmann, S. M. Li, Microbiology 2005, 151, 2199–2207.
[17] S. Maiya, A. Grundmann, S. M. Li, G. Turner, ChemBioChem 2006, 7,
1062–1069.
[18] C. B. Cui, H. Kakeya, G. Okada, R. Onose, H. Osada, J. Antibiot. 1996, 49,
527–533.
[19] H. Woehlecke, H. Osada, A. Herrmann, H. Lage, Int. J. Cancer 2003, 107,
721–728.
[20] L. Wang, K. Sasai, T. Akagi, S. Tanaka, Biochem. Biophys. Res. Commun.
2008, 373, 392–396.
[21] T. Usui, M. Kondoh, C. B. Cui, T. Mayumi, H. Osada, Biochem. J. 1998,
333, 543–548.
[22] H. D. Jain, C. Zhang, S. Zhou, H. Zhou, J. Ma, X. Liu, X. Liao, A. M.
Deveau, C. M. Dieckhaus, M. A. Johnson, K. S. Smith, T. L. Macdonald, H.
Kakeya, H. Osada, J. M. Cook, Bioorg. Med. Chem. 2008, 16, 4626–4651.
[23] E. De Carolis, V. De Luca, Phytochemistry 1994, 36, 1093–1107.
[24] S. G. Sedgwick, S. J. Smerdon, Trends Biochem. Sci. 1999, 24, 311–316.
[25] H. Onaka, S. Asamizu, Y. Igarashi, R. Yoshida, T. Furumai, Biosci. Biotech-
nol. Biochem. 2005, 69, 1753–1759.
[26] W. L. Smith, R. M. Garavito, D. L. DeWitt, J. Biol. Chem. 1996, 271,
33157–33160.
[27] P. J. Punt, R. P. Oliver, M. A. Dingemanse, P. H. Pouwels, C. A. van den
Hondel, Gene 1987, 56, 117–124.
[28] C. B. Cui, H. Kakeya, H. Osada, J. Antibiot. 1996, 49, 534–540.
[29] W.-R. Abraham, H.-A. Arfmann, Phytochemistry 1990, 29, 1025–1026.
[30] S. Kodato, M. Nakagawa, M. Hongu, T. Kawate, T. Hino, Tetrahedron
1988, 44, 359–377.
[31] K. Katayama, K. Masuyama, S. Yoshioka, H. Hasegawa, J. Mitsuhashi, Y.
Sugimoto, Cancer Chemother. Pharmacol. 2007, 60, 789–797.
[32] B. Sarkadi, E. M. Price, R. C. Boucher, U. A. Germann, G. A. Scarborough,
J. Biol. Chem. 1992, 267, 4854–4858.
[33] J. Nakazawa, J. Yajima, T. Usui, M. Ueki, A. Takatsuki, M. Imoto, Y. Y.
Toyoshima, H. Osada, Chem. Biol. 2003, 10, 131–137.
[34] S. Krappmann, C. Sasse, G. H. Braus, Eukaryotic Cell 2006, 5, 212–215.
[35] Y. Ninomiya, K. Suzuki, C. Ishii, H. Inoue, Proc. Natl. Acad. Sci. USA 2004,
101, 12248–12253.
[36] T. Kubodera, N. Yamashita, A. Nishimura, Biosci. Biotechnol. Biochem.
2000, 64, 1416–1421.
[37] S. Takahashi, Y. Zhao, P. E. O’Maille, B. T. Greenhagen, J. P. Noel, R. M.
Coates, J. Chappell, J. Biol. Chem. 2005, 280, 3686–3696.
Received: November 28, 2008
Published online on February 18, 2009
928 www.chembiochem.org  2009 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2009, 10, 920 – 928
H. Osada et al.