Alchivemycin A, a bioactive polycyclic polyketide with an unprecedented skeleton from streptomyces sp.

Article abstract of DOI:10.1021/ol1012982

(Equation Presented). Alchivemycin A, a novel polycyclic polyketide, was isolated from the culture extract of a plant-derived actinomycete Streptomyces sp. The structure and relative configuration were elucidated by spectroscopic analysis and X-ray crystallography, and the absolute configuration was determined by a ¹H NMR anisotropy method using MPA ester derivatization. The new compound contains an unprecedented heterocyclic ring system, 2H-tetrahydro-4,6-dioxo-1,2-oxazine. Alchivemycin A showed potent antimicrobial activity against Micrococcus luteus and inhibitory effects on tumor cell invasion.

Full text of DOI:10.1021/ol1012982

ORGANIC
LETTERS
2010
Vol. 12, No. 15
3402-3405
Alchivemycin A, a Bioactive Polycyclic
Polyketide with an Unprecedented
Skeleton from Streptomyces sp.†
Yasuhiro Igarashi,*,‡ Youngju Kim,‡ Yasuko In,§ Toshimasa Ishida,§
Yukiko Kan,|,⊥ Tsuyoshi Fujita,| Takashi Iwashita,| Hirokazu Tabata,‡
Hiroyasu Onaka,‡ and Tamotsu Furumai‡
Biotechnology Research Center, Toyama Prefectural UniVersity, 5180 Kurokawa,
Imizu, Toyama 939-0398, Japan, Department of Physical Chemistry, Osaka UniVersity
of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki, Osaka 569-1041, Japan, and
Suntory Institute for Bioorganic Research, 1-1-1 Wakayamadai, Shimamoto, Mishima,
Osaka 618-8503, Japan
yas@pu-toyama.ac.jp
Received June 7, 2010
ABSTRACT
Alchivemycin A, a novel polycyclic polyketide, was isolated from the culture extract of a plant-derived actinomycete Streptomyces sp. The
structure and relative configuration were elucidated by spectroscopic analysis and X-ray crystallography, and the absolute configuration was
determined by a 1H NMR anisotropy method using MPA ester derivatization. The new compound contains an unprecedented heterocyclic ring
system, 2H-tetrahydro-4,6-dioxo-1,2-oxazine. Alchivemycin A showed potent antimicrobial activity against Micrococcus luteus and inhibitory
effects on tumor cell invasion.
Actinomycetes have provided a wide array of bioactive
secondary metabolites. In particular, those belonging to the
genus Streptomyces are the most prolific source of chemically
diverse metabolites.2 Their high capacity of secondary
metabolite biosynthesis is largely attributable to the diversity
of polyketide synthase (PKS), which gives rise to more
complex molecular architectures in combination with non-
ribosomal peptide synthase (NRPS).3,4 In our continuing
search for structurally unique metabolites from this genus,5
alchivemycin A (1), a novel metabolite possessing an
unprecedented framework presumably derived from PKS-
† Bioactive microbial metabolites Part 36. For Part 35, see ref 1.
‡ Toyama Prefectural University.
§ Osaka University of Pharmaceutical Sciences.
(2) Be´rdy, J. J. Antibiot. 2005, 58, 1–26
(3) Shen, B. Curr. Opin. Chem. Biol. 2003, 7, 285–295
(4) Dewick, P. M. Medicinal Natural Products-A Biosynthetic Ap-
proach, 3rd ed.; John Wiley & Sons; New York, 2009
.
| Suntory Institute for Bioorganic Research.
.
⊥ Present address: Drug Developmental Research Laboratories, Shionogi
& Co. Ltd., Osaka, Japan.
.
(1) Part 35: Igarashi, Y.; Shimasaki, R.; Miyanaga, S.; Oku, N.; Onaka,
H.; Sakurai, H.; Saiki, I.; Kitani, S.; Nihira, T.; Wimonsiravude, W.;
Panbangred, W. J. Antibiot. 2010, advance online publication June 30, 2010,
DOI: 10.1038/ja.2010.70.
(5) (a) Igarashi, Y.; Iida, T.; Yoshida, R.; Furumai, T. J. Antibiot. 2002,
55, 764–767. (b) Igarashi, Y.; Futamata, K.; Fujita, T.; Sekine, A.; Senda,
H.; Naoki, H.; Furumai, T. J. Antibiot. 2003, 56, 107–113. (c) Igarashi, Y.;
Miura, S.; Fujita, T.; Furumai, T. J. Antibiot. 2006, 59, 193–195.
10.1021/ol1012982  2010 American Chemical Society
Published on Web 07/15/2010

NRPS hybrid biosynthesis, was found from the culture extract
of a Streptomyces strain collected from a leaf of a Chinese
chive (Allium tuberosum)6 by HPLC/UV-based chemical
screening. The producing strain TP-A0867 was cultured in
A-11 M medium at 30 °C for 6 days, and the whole culture
broth was extracted with 1-butanol. UV-guided fractionation
using normal- and reversed-phase column chromatographies,
followed by HPLC purification on a C18 column, yielded
alchivemycin A (1) (Figure 1). The structure and the absolute
stereochemistry of 1 were elucidated by chemical and
spectroscopic methods. The new compound represents the
first example of natural products containing the 2H-tetrahy-
dro-4,6-dioxo-1,2-oxazine ring system.
olefinic, and three oxygen-bearing sp2). The molecular
formula indicated that 1 contained 10 double bond equiva-
lents, which by interpretation of NMR data could be
attributed to two carbon-carbon double bonds (C-2/C-29
and C-26/C-27), two carbonyls (C-1 and C-3), and six ring
systems.
Detailed 2D NMR analysis revealed that 1 was composed
1
of three substructures (Figure 1). From the H-1H COSY
spectrum, five small spin systems were established in
substructure A: H-6/H3-31, H-11/H-12, H-5/H-4/H-13, H3-
32/H-8/H2-9/H-10, and H-15/H2-16/H2-17. These fragments
were connected on the basis of HMBC correlations, leading
to the assembly of the decalin skeleton. The connectivities
from C-5 to C-8 were established by HMBC correlations
from H3-31 to C-5 and C-7 and H3-32 to C-7. Correlations
from H-9 to C-5 and C-11, H-4 to C-10, H-12 to C-4, and
H-13 to C-11 connected C-10 to C-11, C-5 to C-10, and
C-12 to C-13, respectively. Relatively upfield shifts of C-11
and C-12 along with the small coupling constant (3.0 Hz)
between the protons attached to these carbons indicated the
presence of an epoxide ring at these carbons. Additional
small fragment in substructure A could be assembled starting
from the methyl group (C-33). H3-33 showed HMBC
correlations to two oxygenated carbons (C-14 and C-15),
which were assigned to the carbon atoms of an epoxide ring
on the basis of their chemical shifts. This fragment was then
connected to the decalin moiety at C-13 on the basis of
HMBC correlations of H-12 to C-14 and H3-33 to C-13,
thereby establishing substructure A.
Alchivemycin A (1)7 was isolated as optically active
colorless needles ([R]D -17, MeOH; mp 150-153 °C). The
molecular formula of 1 was determined as C35H53NO10 by
high-resolution FABMS data that showed a pseudomolecular
ion at m/z 648.3754 [M + H]+. The IR spectrum of 1
displayed absorption bands at 3433 and 1646 cm-1, indicating
the presence of hydroxyl and carbonyl functionalities,
respectively. The UV spectrum was closely similar to that
reported for tenuazonic acid (λmax 227, 284 nm),8 suggesting
the presence of a tetramic acid-like chromophore in which
an acyl functionality is substituted at the R-position. In the
1H and 13C NMR spectra, most of the signals were doubled
in the ratio of 1.2:1 in CDCl3 and 4:1 in CD3OD. These
observations suggested that 1 existed as a mixture of
tautomeric isomers derived from the above-mentioned con-
jugated system.9 To simplify the analysis, the structure
elucidation was undertaken for the major isomer present in
CD3OD.
Figure 1. 2D NMR correlations for alchivemycin A (1).
1
Interpretation of H, 13C, and HSQC NMR data of 1
In substructure B, three proton-bearing fragments of H-20/
H3-34, H-21/H-22/H-23/H-24/H-25, and H-27/H3-28 were
readily recognized from the 1H-1H COSY spectrum. These
fragments were joined on the basis of HMBC correlations
of H3-34 to C-21, H3-35 to C-25 and C-27, and H3-28 to
C-26. This unit was expanded to include a methylene carbon
C-19 on the basis of an HMBC correlation from H3-34 to
C-19. The carbon chemical shifts suggested the attachment
of oxygen atom to C-21, C-22, C-24, and C-25 and nitrogen
atom to C-23. The geometry of the double bond between
C-26 and C-27 was assigned as E on the basis of an NOESY
correlation between H-25 and H-27, providing substructure
B.
allowed the assignment of signals for 35 carbons including
six methyls, seven methylenes (one is nitrogen- or oxygen-
bearing), 16 methines (eight nitrogen- or oxygen-bearing and
one olefinic), and six quaternary carbons (one sp3, two
(6) Igarashi, Y.; Iida, T.; Sasaki, T.; Saito, N.; Yoshida, R.; Furumai,
T. Actinomycetologica 2002, 16, 9–13.
(7) Alchivemycin A (1): colorless needles; mp 150-153 °C; [R]24D -17
(c 1.0, MeOH); CD (MeOH). ∆ε219 +12.556, ∆ε239 +1.470, ∆ε245 +1.573,
∆ε286 -0.283; UV (MeOH) λmax (log ε) 222 (3.64), 282 (3.85) nm; (0.01
N HCl-MeOH) 222 (3.76), 284 (3.90); (0.01 N NaOH- MeOH) 205 (3.47),
251 (3.92), 274 (3.92); IR νmax 3433, 1646 cm-1. For 1H and 13C NMR
data, see Table S1 in Supporting Information; HRFABMS [M + H]+
648.3754 (calcd for C35H54NO10, 648.3748).
(8) Dictionary of Natural Products on DVD-ROM, Version 18:1;
Chapman & Hall: London, 2009.
Substructure C was elucidated starting from the methylene
protons H2-30 that showed HMBC correlations to C-29, C-2,
(9) (a) Royles, B. J. L. Chem. ReV. 1995, 95, 1981–2001. (b) Schobert,
R.; Schlenk, A. Bioorg. Med. Chem. 2008, 16, 4203–4221.
Org. Lett., Vol. 12, No. 15, 2010
3403

and C-23. HMBC correlations were also observed from H-4
and H-13 to C-3 and from H-4 to C-2. These data established
the four carbon fragment C-3/C-2/C-29/C-30 which was
connected to substructure A between C-3 and C-4 and
substructure B through a nitrogen atom between C-30 and
C-23. The carbon chemical shifts of C-3 (δC 201.3), C-2
(δC 105.3), and C-29 (δC 192.2) could be assigned to the
aforementioned R,ꢀ-unsaturated ketone in which the ꢀ-posi-
tion was oxygenated. Although there was no HMBC cor-
relation available to locate the carbonyl carbon C-1 (δC
176.6), it seemed reasonable that this carbon was connected
to C-2 to extend the conjugation, which was consistent with
the UV spectroscopic data. The remaining oxygen atom,
mandated from the molecular formula, was allowed to be
positioned between C-1 carbonyl carbon and the nitrogen
atom at C-23, constructing the six-membered heterocyclic
ring system. Finally, the remaining methylene fragment (C-
18) was placed between C-17 and C-19, connecting partial
structures A and B to complete the planar structure of 1.
was observed from H-21 to C-29, connecting C-29 and C-22
through an oxygen bridge. The remaining methyl protons
H3-37 and H3-39 were correlated to a quaternary carbon C-36
(δC 113.9) and a carbonyl carbon C-38 (δC 166.4), respec-
tively, establishing the connectivities for C-36/C-37 and
C-38/C-39. Furthermore, four-bond HMBC correlations were
shown from H3-37 to C-3 and H-4 and H3-39 to C-36. These
correlation data, together with its relatively downfield 13C
chemical shift, suggested the connectivity of C-36 to C-3
and C-29 through ether linkages and the attachment of an
acetoxy group to C-36, constructing an orthoacetate. Forma-
tion of this unusual structural unit can be explained by the
following reaction sequence: (1) Michael addition of the
hydroxyl group at C-22 to C-29 and enolization at C-3
carbonyl carbon, (2) acetylation of either 29-OH or 3-OH,
and the following nucleophilic addition of a hydroxyl group
to the acetyl carbonyl (C-36), resulting in the formation of
orthoester-like structure, (3) acetylation of the tertiary
hydroxyl group at C-36.
In the end, because the proposed structure was unusual, 1
was crystallized from a mixture of CH2Cl2-MeOH. This
provided orthorhombic crystals suitable for X-ray analysis,
the result of which confirmed the assigned structure and the
relative configuration of 1 (for crystal structure, see Sup-
porting Information).
The absolute configuration of 1 was determined by the
Trost’s method using R-methoxyphenylacetic acid
(MPA).10,11 Coupling of 1 with 1equiv of (R)- and (S)-MPA
using diisopropylcarbodiimide gave the mono-(R)- and (S)-
MPA esters (3 and 4) as the major products. The esters were
purified by RP-HPLC, and the site of esterification was
confirmed by analysis of 1H, COSY, and HMBC NMR data.
In the H NMR spectra of 3 and 4, positive ∆δR-S (δR -
δS) values were observed for H-21, H-22, H-23, H2-30, and
H3-34, while negative ∆δR-S values were observed for H-27,
H3-28, and H3-35 (Figure 3). These data allowed assignment
of the absolute configuration of C-25 as 25S.
12
Figure 2. Structure and key HMBC correlations of 2.
1
To verify the proposed structure, 1 was treated with acetic
anhydride in pyridine to afford a peracetate (2). The high
resolution FABMS gave an [M + Na]+ peak at m/z 880.4078
appropriate for C45H63NO15 (calcd for C45H63NO15Na m/z
880.4096), corresponding to the addition of five acetate units.
1
In the H NMR spectrum of 2, resonances for five singlet
methyl protons were observed in the region 2.0-2.2 ppm.
The oxymethine protons H-21, H-24, and H-25 were shifted
downfield and had HMBC correlations to the carbonyl
carbons C-40, C-42, and C-44, respectively, to which in turn
correlated singlet methyl protons (H3-41, H3-43, H3-45),
thereby establishing that three hydroxyl groups (21-OH, 24-
OH, 25-OH) were acetylated. The UV spectrum of 2 showed
an absorption maximum at 256 nm, which is similar to that
for actinobolin,8 indicating that the conjugated enone system
was chemically modified during the acetylation. This was
consistent with the 13C NMR data for 2 in which two
deshielded carbon signals for C-3 (δC 201.3) and C-29 (δC
192.2) in 1 were largely shifted upfield to δC 171.3 and δC
89.8, respectively, while the resonance of C-2 carbon was
observed in the similar region. Assignment of these carbons
was established on the basis of HMBC correlations from
H-13 to C-3, H-4 to C-3 and C-2, and H2-30 to C-29 and
C-2 (Figure 2). Furthermore, a four-bond HMBC correlation
Figure 3. ∆δR-S values for mono-MPA derivatives (3 and 4) of
alchivemycin A.
Bioactivity of 1 was evaluated in our standard bioassays
including antimicrobial activity, cytotoxicity, and anti-
(10) (a) Trost, B. M.; Belletire, J. L.; Godleski, S.; McDougal, P. G.;
Balkovec, J. M.; Baldwin, J. J.; Christy, M. E.; Ponticello, G. S.; Varga,
S. L.; Springer, J. P. J. Org. Chem. 1986, 51, 2370–2374. (b) Seco, J. M.;
Quin˜oa´, E.; Riguera, R. Chem. ReV. 2004, 104, 17–117
.
3404
Org. Lett., Vol. 12, No. 15, 2010

invasive activity.13 Compound 1 exhibited a selective
antimicrobial activity against Micrococcus luteus with an
MIC value of 50 nM, whereas it was inactive toward Bacillus
subtilis, Escherichia coli, or Candida albicans. In addition,
1 inhibited the invasion of murine colon carcinoma 26-L5
cells into Matrigel with an IC50 of 0.34 µM without showing
cytotoxic effects and lacked significant cytotoxity against
murine renal carcinoma Renca cells and human umbilical
vein endothelial cells (HUVECs).
incorporated onto the polyketide chain tail, the Dieckmann-
type cyclization provides the 2H-tetrahydro-4,6-dioxo-1,2-
oxazine ring. The 2H-tetrahydro-1,2-oxazine ring system has
been found in natural products from actinomycetes, fungi,
and plants, but none of them are of polyketide origin.15
Figure 4. Plausible biogenesis of 2H-tetrahydro-4,6-dioxo-1,2-
oxazine ring system.
The biosynthetically closest compounds to 1 are the
tetramic acid antibiotics represented by lydicamycins and
delaminomycins from actinomycetes and equisetin-related
fungal metabolites which comprise a decalin unit with a
linear side chain and a tetramic acid unit.16 The most
significant difference of 1 from these compounds is the
additional ring formation between the side chain and the 2H-
tetrahydro-1,2-oxazine ring through a C-N linkage. The
biosynthetic origin of 1 and the structure elucidation of minor
congeners are currently under investigation.
The most structurally intriguing part of 1 is the 2H-
tetrahydro-4,6-dioxo-1,2-oxazine ring system, which has
never been described in either synthetic or natural product
chemistry. We assume that this unique heterocyclic structure
is assembled through a biosynthetic pathway similar to that
for tetramic acid (Figure 4). In polyketide biosynthesis the
tetramic acid moiety is constructed by the condensation of
an R-amino acid with a growing polyketide chain and the
following Dieckmann condensation along with the release
of enzyme complex.9b Similarly, if N-hydroxyglycine, which
may be derived from N-hydroxylation of glycine,14 is
Acknowledgment. We acknowledge Drs. I. Saiki and H.
Sakurai at Toyama University for assistance with anti-
invasion assay. Dr. N. Oku and Mr. T. Fukuda at Toyama
Prefectural University are also thanked for assistance with
NMR measurements.
(11) Mono-(R)-MTPA ester of 1 was obtained by treatment with (S)-
MTPA-Cl, but treatment of 1 with (R)-MTPA-Cl did not afford the (S)-
MTPA ester presumably due to the steric hindrance.
(12) A small amounts of D2O were added to the acetone-d6 solution to
improve the signal broadening due to the keto-enol tautomerization.
(13) (a) Saito, K. I.; Oku, T.; Ata, N.; Miyashiro, H.; Hattori, M.; Saiki,
I. Biol. Pharm. 1997, 20, 345–348. (b) Miyanaga, S.; Obata, T.; Onaka,
H.; Fujita, T.; Saito, N.; Sakurai, H.; Saiki, I.; Furumai, T.; Igarashi, Y. J.
Antibiot. 2006, 59, 698–703.
Supporting Information Available: Experimental details;
1
NMR data and 1D/2D NMR spectra of 1 and 2; H NMR
(14) (a) Thaliath, A.; Socha, D.; Valvano, M. A.; Viswanatha, T. J.
Bacteriol. 1993, 175, 589–596. (b) Parry, R. J.; Li, W.; Cooper, H. N. J.
Bacteriol. 1997, 179, 409–416.
spectra of 3 and 4; ORTEP drawing and a CD spectrum of
1; X-ray crystallographic file in CIF format for 1. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(15) (a) Kiyoto, S.; Shibata, T.; Yamashita, M.; Komori, T.; Okuhara,
M.; Terano, H.; Kohsaka, M.; Aoki, H.; Imanaka, H. J. Antibiot. 1987, 40,
594–599. (b) Fujita, T.; Takase, S.; Otsuka, T.; Terano, H.; Kohsaka, M. J.
Antibiot. 1988, 41, 392–394. (c) Terano, H.; Takase, S.; Hosoda, J.; Kohsaka,
M. J. Antibiot. 1989, 42, 145–148. (d) Hootele´, C. Tetrahedron Lett. 1969,
2713, 2716. (e) Stipanovic, R. D.; Howell, C. R. J. Antibiot. 1982, 35, 1326–
1330. (f) Kanai, Y.; Fujimaki, T.; Kochi, S.; Konno, H.; Kanazawa, S.;
Tokumasu, S. J. Antibiot. 2004, 57, 24–28. (g) Capon, R. J.; Ratnayake,
R.; Stewart, M.; Lacey, E.; Tennant, S.; Gill, J. H. Org. Biomol. Chem.
2005, 3, 123–129. (h) Horii, Z.; Imanishi, T.; Yamauchi, M.; Hanaoka,
M.; Parello, J.; Munavalli, S. Tetrahedron Lett. 1972, 13, 1877–1880. (i)
Yu, R.; Li, B. G.; Ye, Q.; Zhang, G. L. Nat. Prod. Res. 2005, 19, 359–362.
(j) Ohsaki, A.; Kobayashi, Y.; Yoneda, K.; Kishida, A.; Ishiyama, H. J.
Nat. Prod. 2007, 70, 2003–2005.
OL1012982
(16) (a) Hayakawa, Y.; Kanamaru, N.; Morisaki, N.; Furihata, K.; Seto,
H. J. Antibiot. 1991, 44, 288–292. (b) Furumai, T.; Eto, K.; Sasaki, T.;
Higuchi, H.; Onaka, H.; Saito, N.; Fujita, T.; Naoki, H.; Igarashi, Y. J.
Antibiot. 2002, 55, 873–880. (c) Ueno, M.; Someno, T.; Sawa, R.; Iinuma,
H.; Naganawa, H.; Ishizuka, M.; Takeuchi, T. J. Antibiot. 1993, 46, 979–
984. Vesonder, R. F.; Tjarks, L. W.; Rohwedder, W. K.; Burmeister, H. R.;
Laugal, J. A. J. Antibiot. 1979, 32, 759–761. Marfori, E. C.; Kajiyama, S.;
Fukusaki, E.; Kobayashi, A. Z. Naturforsch. C 2002, 57, 465–470.
Org. Lett., Vol. 12, No. 15, 2010
3405