Unprecedented biological cyclopropanation in the biosynthesis of FR-900848

Article abstract of DOI:10.1039/b809610d

We were able to show the predominant incorporation of a single enantiomer and intact incorporation of multiply labelled synthetic diketide precursors (14 and 16), which established the intermediacy of cyclopropanated diketide and led to our proposal for t

Full text of DOI:10.1039/b809610d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Unprecedented biological cyclopropanation in the biosynthesis
of FR-900848w
Tetsuo Tokiwano,*b Hiroaki Watanabe,a Takashi Seoa and Hideaki Oikawa*a
Received (in Cambridge, UK) 6th June 2008, Accepted 12th September 2008
First published as an Advance Article on the web 14th October 2008
DOI: 10.1039/b809610d
We were able to show the predominant incorporation of a single
enantiomer and intact incorporation of multiply labelled syn-
thetic diketide precursors (14 and 16), which established the
intermediacy of cyclopropanated diketide and led to our propo-
sal for the unprecedented biological cyclopropanation, via PKS
(polyketide synthase) having a novel cyclopropanase domain, in
the biosynthesis of FR-900848 (1).
FR-900848 (1) is
a polyketide-nucleoside produced by
Streptoverticillium fervens HP-891 and shows potent activity
against phytopathogenic fungi.1 Its structure consists of a
cyclopropanated fatty acid containing one isolated and four
contiguous cyclopropanes,2 and 500-amino-500-deoxy-50,60-di-
hydrouridine (Fig. 1). The closely related compound
U-106305 (3), an inhibitor of the cholesteryl ester transfer
protein (CETP), was isolated from Streptomyces sp. UC 11136
by Kuo et al.3 and was found to have the same absolute
stereochemistry for cyclopropane rings.4 To elucidate the
mechanism for enzymatic construction of polycyclopropanes,
we have started on a biosynthetic study of FR-900848 (1) and
determined its biosynthetic building units by a series of feeding
experiments with isotopically labelled precursors.5a,b These
results established that the backbone of 1 was constructed
via a polyketide pathway and was coupled with an amino-
nucleoside unit derived from dihydrouridine. In addition,
methylenes of the most characteristic cyclopropane moieties
in 1 were derived from L-methionine.
Fig. 1
chain (Scheme 1, path b).4 Herein we describe the incorpora-
tion of plausible cyclopropanated diketide precursors into 1,
and we further propose an unprecedented biological cyclo-
propanation involved in the biosynthesis of FR-900848.
To examine the mechanism of the step-wise introduction of
the cyclopropane system (Scheme 1, path b), we planned a
feeding experiment with several cyclopropanated diketide
precursors labelled with deuterium. Incorporation of
advanced intermediates as thioester-activated forms is a well
The introduction of the highly unusual contiguous cyclo-
propane moieties is the most remarkable step in the biosynth-
esis of FR-900848 (1). For the intriguing cyclopropanations,
two biosynthetic pathways were proposed as depicted in
Scheme 1. Kuo et al. proposed that in the biosynthesis of
U-106305 the octaenoic acid 6, which is constructed via the
polyketide pathway, is cyclopropanated by methyl transfer
from S-adenosylmethionine (SAM), followed by ring closure
(Scheme 1, path a).3 Alternatively, based on the well-known
cyclopropanation of the a,b-unsaturated carbonyl system with
sulfur-ylide,6 Barrett et al. proposed a biosynthetic pathway in
which cyclopropanation occurs in the growing the polyketide
a Division of Chemistry, Graduate School of Science, Hokkaido
University, Sapporo, 060-0810, Japan.
E-mail: hoik@sci.hokudai.ac.jp
b Department of Biotechnology, Faculty of Bioresource Sciences, Akita
Prefectural University, Akita, 010-0195, Japan.
E-mail: tkwn@akita-pu.ac.jp
w Electronic supplementary information (ESI) available: Experimental
details and NMR spectra. See DOI: 10.1039/b809610d
Scheme 1 Two proposed biosynthetic pathways to the polycyclo-
propane unit of 1.
ꢀc
This journal is The Royal Society of Chemistry 2008
6016 | Chem. Commun., 2008, 6016–6018

View Article Online
Scheme 2 Reagents and conditions: (i) TBDPSCl, imidazole, DMF,
25 1C; (ii) BuLi, CD3OTs, THF, 25 1C; (iii) HF–pyridine, THF, 25 1C;
(iv) LiAlH4, NaOMe, THF, ꢁ10 1C to reflux; (v) CH2I2, Et2Zn,
CH2Cl2, 0 1C; (vi) RuCl3, NaIO4, CCl4–H2O–CH3CN, 25 1C; (vii)
N-acetylcysteamine, DCC, Ph3P, DMAP, CH2Cl2, 0 1C; (viii) LiAlH4,
NaOMe, THF, ꢁ10 1C to reflux, then D2O.
established strategy in biosynthetic studies of polyketides.7a
Synthesis of the plausible precursor 14 and its enantiomer
(ent-14) is shown in Scheme 2. The lithium acetylide prepared
from the corresponding silyl ether 9 was reacted with
C2H3OTs to afford 10 in 95% yield (499 atom% 2H).
After deprotection of the silyl ether 10, hydroalumination of
the resulting 2-butyn-1-ol with LiAlH4–NaOMe yielded
[4,4,4-2H3]-2-buten-1-ol 11. Cyclopropanation of 11 using
Et2Zn–CH2I2 in the presence of D-tartramide-derived dioxa-
borolane (12) according to the Charette protocol8 gave 13. The
optical purity of 13 was determined to be 89% ee by 1H-NMR
analysis of the corresponding (R)-MTPA ester. Oxidation of
the alcohol 13 with RuCl3–NaIO4 followed by condensation
with N-acetyl cysteamine using DCC, Ph3P, and DMAP
afforded the thioester 14. Charette cyclopropanation with
the enantiomeric dioxaborolane afforded ent-13 (91% ee)
which was converted to ent-14 in the same procedure. To
confirm the intact incorporation, the multiply deuterium
labelled diketide precursor 16 was synthesized. Hydroalumi-
nation of [4,4,4-2H3]-2-butynl-1-ol followed by quenching with
2H2O provided a mixture of [3,4,4,4-2H4]-2-buten-1-ol (15)
and [2,4,4,4-2H4]-2-buten-1-ol (150). The labelled 2-buten-1-ols
(15 and 150) were converted to 16 and 160 as a 65 : 35 mixture9
according to the same procedure as that to obtain 14.
Fig. 2 (a) 1H NMR spectrum of 2; (b, c, d) 2H NMR spectra of 2
derived from the feeding experiments with 14, ent-14, and 16 (160),
respectively.
observed (Fig. 2, c). Incorporations of 14 and ent-14 into 2
were 0.11% and 0.01%, respectively, as determined with
reference to the natural abundance solvent (CHCl3) peaks.
Considering the optical purity of ent-14, the signal is likely
derived from a small amount of 14. To obtain further support
for these experimental results, the incorporation of multiply
2
deuterium labelled 16 and 160 was employed. The H-NMR
spectrum of 2 from the feeding experiment with the mixture of
16 and 160 showed signals of 18-C2H3, which overlapped with
16-2H, and 17-2H at 1.02 and 0.67 ppm, respectively (Fig. 2,
d), in the integral ratio of 6.8 : 1 which is nearly identical to the
corresponding ratio, 5.2 : 1 [(H4 + H3) : (H2)] in 16 and 160.
These results strongly indicated the specific and intact
incorporation of the precursors 14 and 16 into 1 without
degradation in a metabolic pathway.
Separate administrations of synthetic analogues 14 and
ent-14 to S. fervens HP-891 followed by the work-up proce-
dure described previously afforded diacetate 2 which was
subjected to NMR analysis.5a The 2H-NMR spectrum of 2
derived from the feeding experiment with 14 having the same
absolute stereochemistry as that of FR-900848, showed the
signal at 1.02 ppm corresponding to 18-Me (Fig. 2, b). In the
2H-NMR spectrum of 2 provided by the administration of
ent-14, the 18-C2H3 signal with very low intensity was
Since we have already established that the backbone of 1 is
biosynthesized via a polyketide pathway, probably by putative
type-I PKS,5a predominant incorporation of the chiral dike-
tide precursor 14 strongly indicated that cyclopropanation
occurred at the stage of a,b-unsaturated thioester bound to
PKS. In the biosynthesis of cyclopropanated natural products,
such as bacterial lipids, mycolic acids10 and sterols,11 the
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 6016–6018 | 6017

View Article Online
We are grateful to Dr Hideyuki Muramatsu of Astellas
Pharmaceutical Co. Ltd for supplying Streptoverticillium
fervens HP-891 and to the GC-MS and NMR Laboratory,
Graduate School of Agriculture, Hokkaido University, for the
NMR measurements. This study was supported financially by
the Grants-in-Aid for Scientific Research [(A)17208010 to H.
Oikawa and 19710174 to T. Tokiwano] from the Japan Society
for the Promotion of Science.
Notes and references
Scheme 3 Proposed cyclopropanation with PKS having a cyclo-
propanase domain.
1 M. Yoshida, M. Ezaki, M. Hashimoto, M. Yamashita, N.
Shigematsu, M. Okuhara, M. Kohsaka and K. Horikoshi,
J. Antibiot., 1990, 43, 748–754.
cyclopropane moiety is introduced to the isolated olefin with
SAM and a corresponding enzyme. However, it is less likely
that the electrophile SAM directly methylates the electron
deficient olefin 4 to give the cyclopropanated diketide 5.
On the other hand, the alternative cyclopropanation of the
a,b-unsaturated carbonyl system with sulfur-ylide (Scheme 3)
fits well with our experimental results. Considering the
involvement of type-I PKS, which consists of common
domains,7a,b we propose an additional domain, cyclopropanase
(CP), which catalyzes the conjugate addition of the
SAM-derived ylide to the a,b-unsaturated thioester, followed
by ring closure of the resulting enolate intermediate (Scheme 3,
A) thus affording the cyclopropane ring. This domain could be
regarded as a derivative of the methylation domain which is
found in the bacterial type-I PKS.12a,b In the putative
FR-900848 PKS, the 1st and 3rd–6th modules could have CP
domains responsible for cyclopropanation in the corresponding
position.
2 A. G. M. Barrett and K. Kasdorf, J. Am. Chem. Soc., 1996, 118,
11030–11037.
3 M. S. Kuo, R. J. Zielinski, J. I. Cialdella, C. K. Marschke, M. J.
Dupuis, G. P. Li, D. A. Kloosterman, C. H. Spilman and V. P.
Marshall, J. Am. Chem. Soc., 1995, 117, 10629–10634.
4 A. G. M. Barrett, D. Hamprecht, A. J. P. White and D. J.
Williams, J. Am. Chem. Soc., 1997, 119, 8608–8615.
5 (a) H. Watanabe, T. Tokiwano and H. Oikawa, Tetrahedron Lett.,
2006, 47, 1399–1402; (b) H. Watanabe, T. Tokiwano and H.
Oikawa, J. Antibiot., 2006, 59, 607–610.
6 E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 1965, 87,
1353–1364.
7 (a) B. J. Rawlings, Nat. Prod. Rep., 2001, 18, 190–227; (b) M. A.
Fischbach and C. T. Walsh, Chem. Rev., 2006, 106, 3468–3496.
8 A. B. Charette and H. Juteau, J. Am. Chem. Soc., 1994, 116,
2651–2652.
9 The isotopic enrichment (86 atom% 2H) and the ratio of 16 and 160
were determined by 1H-NMR analysis. The optical purity (91% ee)
was determined by H-NMR analysis of the (R)-MTPA ester 17.
1
In conclusion, our experimental results on intact incorpora-
tion of multiply labelled biosynthetic precursor analogues
established the intermediacy of the cyclopropanated diketide
and suggested the unprecedented biological cyclopropanation
10 D. F. Iwig, A. Uchida, J. A. Stromberg and S. J. Booker, J. Am.
Chem. Soc., 2005, 127, 11612–11613.
11 L. A. Wessjohann, W. Brandt and T. Thiemann, Chem. Rev., 2003,
103, 1625–1647.
with PKS, having novel
a cyclopropanase domain, in
12 (a) K. Arakawa, F. Sugino, K. Kodama, T. Ishii and H. Kinashi,
Chem. Biol., 2005, 12, 249–256; (b) L. Tang, S. Shah, L. Chung, J.
Carney, L. Katz, C. Khosla and B. Julien, Science, 2000, 287,
640–642.
the biosynthesis of FR-900848 (1). Based on the pathway
proposed in this paper, we are currently identifying the
biosynthetic gene cluster.
ꢀc
This journal is The Royal Society of Chemistry 2008
6018 | Chem. Commun., 2008, 6016–6018