Synthesis of the (-)-TAN-2483B ring system via a D-mannose-derived cyclopropane

Article abstract of DOI:10.1039/c0ob00851f

The ring system of the fungal metabolite (-)-TAN-2483B has been synthesised, for the first time, from d-mannose, utilising a cyclopropanation/ring expansion sequence.

Full text of DOI:10.1039/c0ob00851f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 998
www.rsc.org/obc
COMMUNICATION
Synthesis of the (-)-TAN-2483B ring system via a D-mannose-derived
cyclopropane†
Russell J. Hewitt and Joanne E. Harvey*
Received 8th October 2010, Accepted 26th November 2010
DOI: 10.1039/c0ob00851f
The ring system of the fungal metabolite (-)-TAN-2483B has
been synthesised, for the first time, from D-mannose, utilising
a cyclopropanation/ring expansion sequence.
the ring system of (-)-TAN-2483B (2), accessed from D-mannose
through a key cyclopropanation/ring expansion sequence.
(-)-TAN-2483B possesses multiple chiral centres and several
oxygen groups. The stereochemical diversity contained within
the carbohydrate chiral pool provided a suitable scaffold for
this synthetic endeavour, wherein the ring expansion of a gem-
dihalocyclopropane7 fused to a five-membered ring glycal would
generate the ring unsaturation. We envisaged that (-)-TAN-2483B
could be accessed from furanopyranone 3 (Scheme 1), which
in turn could be derived from hemiacetal 4 through Wittig
homologation and lactone formation. This substrate could be
prepared by gem-dihalocyclopropanation of known glycal 5,8
followed by silver-promoted ring expansion. The synthesis of a
simplified furanopyrone, such as 6, would provide a proof-of-
principle for the proposed approach to the natural product, and is
described herein.
Bioactive natural products incorporating the furo[3,4-b]pyran-5-
one bicyclic system have been isolated from a variety of fungal
sources. Fusidilactones A, B, D and E,1 massarilactones B and
D,2 TAN-2483A and B3 and waol A4,5 all contain this ring
system. These fungal secondary metabolites display a variety of
bioactivities, ranging from antibacterial to anti-tumour properties.
To date, only the Snider group has reported the synthesis of
members of this family of natural products,5 namely (-)-TAN-
2483A (1), massarilactone B and waol A (Fig. 1).6 These natural
products either incorporate a degree of unsaturation across the
fused 4a–7a bond (e.g. massarilactone B) or possess a cis-
relationship between H-2 and H-7a (e.g. (-)-TAN-2483A, waol A).
In contrast to the other members of this family, (-)-TAN-2483B
(2), isolated from a Japanese filamentous fungus,3 has a trans-
relationship between H-2 and H-7a, and was inaccessible through
the reported5 methodology. Herein, we present the first synthesis of
Scheme 1 Retrosynthetic analysis of TAN-2483B.
Glycal 5 was prepared from D-mannose using the four-step
sequence described by Theodorakis.8 Initial attempts to cyclo-
propanate the alkene moiety of 5 with dibromocarbene gave
complex mixtures in which the desired cyclopropane was not
evident. In contrast, addition of dichlorocarbene to 5 under
Ma˛kosza conditions9 was a facile process, cleanly yielding a single
isomer of cyclopropane 7 (Scheme 2). The propensity of this
product to undergo spontaneous ring expansion led to its isolation
in variable yields (50–79%) after column chromatography. Ring-
expansion of cyclopropane 7 was achieved by treatment with
silver acetate and sodium acetate,10 and provided a mixture of
Fig. 1 Selected furo[3,4-b]pyran-5-one fungal natural products.
School of Chemical and Physical Sciences, Victoria University of Wellington,
PO Box 600, Wellington, New Zealand. E-mail: joanne.harvey@vuw.ac.nz;
Fax: +64 4 4635237; Tel: +64 4 4635956
† Electronic supplementary information (ESI) available: Experimental
procedures and characterisation data for all new compounds. See DOI:
10.1039/c0ob00851f
998 | Org. Biomol. Chem., 2011, 9, 998–1000
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
of (-)-TAN-2483B. Only the desired isomer of6 was obtained (18%
yield), in addition to recovered 11 (63% yield), which was enriched
in 11a. It appears that, surprisingly, the major isomer of the
alkenyl chloride, 11a, is unreactive towards carbonylation under
the conditions used. The stereochemical assignment of 6 was based
on nOe experiments (Fig. 2). The lack of a significant nOe between
H-7a and H-2 provided circumstantial evidence for a trans-
relationship between H-2 and H-7a across the ring,11 indicating
that the stereochemistry was 7aS, as in (-)-TAN-2483B. Future
synthetic efforts will initially centre around improving the overall
yield of 6 by augmenting the selectivity of the epoxidation and
exploring alternative methods for epoxide opening. Our strategy
provides suitable functional handles for synthesis of the natural
product itself. For instance, removal of the acetonide protecting
group in furo[3,4-b]pyran-5-one 6, followed by diol cleavage,
Julia–Kocienski olefination and benzyl ether deprotection would
produce desmethyl (-)-TAN-2483B. Furthermore, incorporation
of the C-7 methyl group, either into an oxidised derivative of
alcohol 11 or during the homologation of 4, would provide the
natural product. Our work to these ends will be reported in due
course.
Scheme 2 Synthesis of the pyran ring via a cyclopropane.
anomeric acetates 8 (49% combined yield), in addition to the
corresponding 3-acetylglycal (16%), formed by acetate attack at
C-2 of cyclopropane 7. Gratifyingly, we found that the desired
product (8) could be obtained in much higher yield (77% over
two steps) by treating the crude cyclopropane 7 with silver acetate
in acetic acid.‡ The mixture of acetates was then subjected to
methanolysis to afford 4 in a quantitative yield.
Construction of the lactone ring was initiated by Wittig
olefination of the masked aldehyde 4 using the ylide derived
from methyltriphenylphosphonium bromide, producing diene 9
(Scheme 3). Epoxidation of this diene using Jacobsen’s cata-
lyst provided very low conversion, with poor stereoselectivity.
Encouragingly, epoxidation with m-CPBA produced a mixture
of stereoisomers 10 in a 1.5:1 ratio (52% yield). Base-mediated
intramolecular epoxide opening of 10 with sodium hydride in
DMF was sluggish, affording the product 11 in 23% yield (35%
based on recovered starting material) and as a 1.5:1 ratio of
epimers, which correlated with the mixture of isomers subjected
to this process. Reaction of 10 with methanolic sodium methoxide
provided the desired functionalised pyran 11 in a better yield (42%)
but as a 3.3:1 mixture of epimers. In this case, the product 11
was accompanied by two by-products, thought to be the seven-
membered ring regioisomer and a methyl ether formed by attack of
methoxide on the epoxide. The assignment of the major and minor
isomers of 11 from these reactions as 11a and 11b, respectively, was
made after lactone formation (vide infra).
Fig. 2 Key nOe results used in the stereochemical assignment of 6.
The route described above represents, to our knowledge, the
first synthesis of the (-)-TAN-2483B ring system, accessed from
D-mannose in 11 steps. This strategy provides scope for inclusion
of the functionality present in the natural product.
Acknowledgements
The authors thank the Cancer Society of New Zealand for a
PhD Scholarship (RJH) and Victoria University of Wellington for
funding. We are grateful to Prof. John Spencer and his group for
assistance with the carbonylation, and to Dr John Ryan, Jonathan
Singh, Dr Robert Keyzers and A. Prof. Peter Northcote for their
advice and help with NMR spectroscopy of the final compound.
We acknowledge A. Prof. Paul Teesdale-Spittle and Em. Prof.
Brian Halton for helpful discussions throughout this work.
Notes and references
‡ Acetate 8. Glycal 5 (462 mg, 1.67 mmol) was dissolved in chloroform
(3 mL), treated with TEBAC (8 mg, 0.04 mmol), followed by a solution of
sodium hydroxide (3.01 g, 75.3 mmol) in water (3.0 mL) and allowed
to stir at room temperature for 2.5 hours. The reaction mixture was
dissolved in water, then extracted with diethyl ether. The combined ethereal
fractions were dried, filtered and concentrated to afford a brown oil. This
oil was passed through a plug of silica gel and eluted with 1:5 ethyl
acetate/hexanes to afford crude cyclopropane 7 as a yellow oil. This
sample was dissolved in acetic acid (17 mL), treated with silver acetate
(435 mg, 2.61 mmol), and then stirred at 80 ◦C for two hours. The solution
was filtered through CeliteꢀR and eluted with diethyl ether. The reaction
mixture was concentrated to provide a yellow oil, which was purified by
Scheme 3 Formation of the TAN-2483B ring system.
The mixture of inseparable alcohols 11a and 11b was subjected
to a palladium-catalysed carbonylation/lactonisation sequence
under an atmosphere of carbon monoxide. This provided the
furo[3,4-b]pyran-5-one 6,§ incorporating the bicyclic ring system
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 998–1000 | 999
©

View Article Online
column chromatography (1:5 EtOAc/hexanes), affording an inseparable
mixture of acetyl pyranoside anomers 8 (495 mg, 77% over two steps, 3:1
ratio) as a pale-yellow oil.
2 (a) H. Oh, D. C. Swenson, J. B. Gloer and C. A. Shearer, Tetrahedron
Lett., 2001, 42, 975–977; (b) I. Kock, K. Krohn, H. Egold, S. Draeger, B.
Schulz and J. Rheinheimer, Eur. J. Org. Chem., 2007, 2186–2190; (c) K.
Krohn, Z. Ullah, H. Hussain, H., U. Flo¨rke, B. Schulz, S. Draeger,
G. Pescitelli, P. Salvadori, S. Antus and T. Kurta´n, Chirality, 2007, 19,
464–470.
§ Furo[3,4-b]pyran-5-one 6. A solution of alcohols 11a and 11b (40 mg,
0.11 mmol, 2.8:1 ratio) in 1,4-dioxane (3.0 mL) was treated with
palladium(II) acetate (10 mg, 0.044 mmol), triphenylphosphine (60 mg,
0.23 mmol) and N,N-diisopropylethylamine (30 mL, 0.17 mmol), then
stirred at reflux under an atmosphere of carbon monoxide for one day.
The solution was filtered through a pad of silica, then washed with diethyl
ether. The ethereal solution was dried, filtered and concentrated to provide
a brown oil. Upon column chromatography (1:3 EtOAc/hexanes), starting
material 11 was recovered (3.7:1 ratio, 25 mg, 63%) and bicycle 6 (7 mg,
18%) was obtained as a pale-yellow solid. Rf 0.3 (1:3 EtOAc/hexanes); [a]D18
-142 (c 0.5, CHCl3); 1H NMR (500 MHz, CDCl3) d 7.37–7.31 (complex
m, 5H, Bn), 7.25 (dd, J = 6.1, 3.4 Hz, 1H, H-4), 5.18 (app. td, J = 7.9,
3.0 Hz, 1H, H-7a), 4.64 (d, J = 11.5 Hz, 1H, one of PhCH2), 4.60 (partially
obscured dd, J = 8.7, 8.3 Hz, 1H, one of H-7), 4.59 (d, J = 11.5 Hz, 1H,
one of PhCH2), 4.42–4.38 (complex m, 2H, H-3 and H-4¢), 4.13 (dd, J =
8.6, 6.2 Hz, 1H, one of H-5¢), 4.00 (dd, J = 8.8, 7.6 Hz, 1H, one of H-7),
3.96 (dd, J = 8.8, 5.1 Hz, 1H, one of H-5¢), 3.38 (dd, J = 8.3, 0.9 Hz, 1H,
H-2), 1.38 (s, 6H, (CH3)2C); 13C NMR (125 MHz, CDCl3) d 167.0 (C,
C-5), 137.6 (C, Ph), 135.4 (CH, C-4), 133.8 (CH, C-4a), 128.5 (CH, Ph),
128.0 (CH, Ph), 127.8 (CH, Ph), 109.2 (C, (CH3)2C), 75.7 (CH, C-2), 73.5
(CH, C-4¢), 72.0 (CH2, PhCH2), 71.1 (CH2, C-7), 70.3 (CH, C-7a), 67.5
(CH2, C-5¢), 67.3 (CH, C-3), 26.9 (CH3, one of (CH3)2C), 25.3 (CH3, one
of (CH3)2C); IR (KBr): 2984, 2923, 2874, 1772, 1377, 1369, 1204, 1108,
1067, 768 cm-1; HRMS (ESI) calcd for C19H22O6Na+ [M + Na]+ 369.1314,
found 369.1308.
3 K. Hayashi, M. Takizawa, and K. Noguchi, Jpn. Patent 10287679,
1998; Chem. Abstr. 1999, 130, 3122e.
4 (a) O. Nozawa, T. Okazaki, N. Sakai, T. Komurasaki, K. Hanada, S.
Morimoto, Z.-X. Chen, B.-M. He and K. Mizoue, J. Antibiot., 1995,
48, 113–118; (b) O. Nozawa, T. Okazaki, S. Morimoto, Z.-X. Chen,
B.-M. He and K. Mizoue, J. Antibiot., 2000, 53, 1296–1300.
5 (a) X. Gao, M. Nakadai and B. B. Snider, Org. Lett., 2003, 5, 451–454;
(b) X. Gao and B. B. Snider, J. Org. Chem., 2004, 69, 5517–5527.
6 A multi-component approach to the furo[3,4-b]pyran-5-one ring system
has also been reported: A. Shaabani, E. Soleimani, A. Sarvary and A.
H. Rezayan, Bioorg. Med. Chem. Lett., 2008, 18, 3968–3970.
7 (a) M. G. Banwell, Pure Appl. Chem., 2008, 80, 669–679; (b) B. Halton
and J. Harvey, Synlett, 2006, 1975–2000; (c) M. Fedoryn´ski, Chem.
Rev., 2003, 103, 1099–1132.
8 C. Kim, R. Hoang and E. A. Theodorakis, Org. Lett., 1999, 1, 1295–
1297.
9 M. Ma˛kosza and M. Wawrzyniewicz, Tetrahedron Lett., 1969, 10, 4659–
4662.
10 R. J. Hewitt and J. E. Harvey, J. Org. Chem., 2010, 75, 955–958.
11 Molecular mechanics calculations (see Supporting Information) indi-
cate that the distance between H-2 and H-7a in the undesired epimer,
˚
7a-epi-6, would be ca. 2.38 A, so a strong nOe signal would be expected.
The absence of an nOe implies that the desired isomer, 6, is formed.
This assignment is corroborated by nOe enhancements between H-7a
and protons on the top face as drawn: H5¢, an aromatic proton of
the benzyl group, and the proton attached to C-7 that does not interact
with H-2 (the other H-7 displays a weaker nOe with H-7a and a distinct
correlation with H-2).
1 (a) K. Krohn, C. Biele, K.-H. Drogies, K. Steingro¨ver, H.-J. Aust, S.
Draeger and B. Schulz, Eur. J. Org. Chem., 2002, 2331–2336; (b) S.
Qin, K. Krohn, U. Flo¨rke, B. Schulz, S. Draeger, G. Pescitelli, P.
Salvadori, S. Antus and T. Kurta´n, Eur. J. Org. Chem., 2009, 3279–
3284.
1000 | Org. Biomol. Chem., 2011, 9, 998–1000
This journal is
The Royal Society of Chemistry 2011
©