Bioinspired synthesis of hirsutellones A, B, and C

Article abstract of DOI:10.1021/ol202239u

The total synthesis of hirsutellones A (1), B (2), and C (3) has been achieved through a bioinspired late-stage sequence starting from advanced intermediate 6. The sequence proceeded via labile intermediate 17,1′-dehydrohirsutellone B (5) and delivered, i

Full text of DOI:10.1021/ol202239u

ORGANIC
LETTERS
2011
Vol. 13, No. 20
5708–5710
Bioinspired Synthesis of Hirsutellones
A, B, and C
K. C. Nicolaou,*,†,‡ Ya-Ping Sun,† David Sarlah,† Weiqiang Zhan,† and T. Robert Wu†
Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps
Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, United
States, and Department of Chemistry and Biochemistry, University of California, San
Diego, 9500 Gilman Drive, La Jolla, California 92093, United States
kcn@scripps.edu
Received August 18, 2011
ABSTRACT
The total synthesis of hirsutellones A (1), B (2), and C (3) has been achieved through a bioinspired late-stage sequence starting from advanced
intermediate 6. The sequence proceeded via labile intermediate 17,10-dehydrohirsutellone B (5) and delivered, in addition to the natural products
(1ꢀ3), hirsutellone analogue 10,20,17-epi-hirsutellone C (10,20,17-epi-3).
Heterodimeric hirsutellone F (4) and 17,10-dehydrohir-
sutellone B (5),2 the putative biosynthetic precursor to
hirsutellones 1ꢀ4 (Scheme 1), were recently added to the
growing class of hirsutellones [A (1), B (2), and C (3),1
Scheme 1]. Isolated from the seed fungus Trichoderma sp.
BCC 7579 incubated in a bioreactor (as opposed to
incubation in an Erlenmeyer flask) by Isaka and co-work-
ers, hirsutellone F (4) was shown to decompose to its
apparent components, 17,10-dehydrohirsutellone B (5)
and hirsutellone A (1), upon exposure to basic conditions
as shown in Scheme 1.2 When the decomposition of 4 was
carried out in the presence of NaBH4, hirsutellone B (2)
was obtained in addition to hirsutellone A (1). Further-
more, it was determined that when the basic decomposi-
tion of 4 was performed in the presence of H2O2ꢀNaOH,
hirsutellone C (3) was generated together with hirsutellone
A (1). Apparently, in the latter reactions, the fleeting
intermediate 5 was intercepted (17,10-reduction or epoxi-
dation) to a significant extent prior to its conversion to
hirsutellone A (1) through based-induced bond migra-
tion accompanied by ring expansion. These observations
prompted the isolation chemists to propose a late-stage
biosynthetic hypothesis involving labile intermediate 5 as
the biogenetic precursor of all hirsutellones shown in
Scheme 1.3 In this communication, we report the in situ
generation of this 17,10-dehydrohirsutellone B (5) and its
conversion to hirsutellones A (1), B (2), and C (3), thereby
confirming the Isaka hypothesis and achieving the first
total synthesis of hirsutellones A (1) and C (3) and a second
generation total synthesis of hirsutellone B (2).4,5 These
natural products are notable not only for their novel
molecular architectures but also for their promising anti-
tuberculosis properties (active against Mycobacterium
tuberculosis, MIC = 0.78 μg/mL for 1ꢀ3).1
Our initial attempts to obtain the desired 17,10-dehydro-
hirsutellone B (5) through the obvious route involving site-
selective oxidation of hirsutellone B (2) failed, presumably
due to steric hindrance at C-17 and the arrangement of the
β-ketoamide moiety carbonyl groups of the substrate (2)
(3) A similar late-stage biosynthetic hypothesis has been reported
previously in a series of GKK1032 derivatives; see: Oikawa, H. J. Org.
Chem. 2003, 68, 3552–3557.
(4) Nicolaou, K. C.; Sarlah, D.; Wu, T. R.; Zhan, W. Angew. Chem.,
Int. Ed. 2009, 48, 6870–6874.
(5) For previous synthetic studies toward hirsutellones, see: (a) Tilley,
S. D.; Reber, K. P.; Sorensen, E. J. Org. Lett. 2009, 11, 701–703. (b)
Huang, M.; Huang, C.; Liu, B. Tetrahedron Lett. 2009, 50, 2797–2800.
(c) Huang, M.; Song, L.; Liu, B. Org. Lett. 2010, 12, 2504–2507. (d)
Halvorsena, G. T.; Roush, W. R. Tetrahedron Lett. 2011, 52, 2072–2075.
† The Scripps Research Institute.
‡ University of California, San Deigo.
(2) Isaka, M.; Prathumpai, W.; Wongsa, P.; Tanticharoen, M. Org.
Lett. 2006, 8, 2815–2817.
(1) Isaka, M.; Rugseree, N.; Maithip, P.; Kongsaeree, P.; Prabpai, S.;
Thebtaranonth, Y. Tetrahedron 2005, 61, 5577–5583.
r
10.1021/ol202239u
Published on Web 09/14/2011
2011 American Chemical Society

Scheme 1. Molecular Structures of Hirsutellones A (1), B (2), C
(3), and F (4), Their Postulated Biogenetic Precursor 17,10-
Dehydrohirsutellone B (5), and Isaka’s Confirmed Late-Stage
Biosynthetic Hypothesis for Hirsutellones 1ꢀ32
Scheme 2. Synthesis of Keto Furanone 9
Scheme 3. Preparation of Hydroxyamide 12
(most likely locked in an eclipsed conformation). Finding
ourselves in this predicament, we decided to explore the
development of a synthetic strategy to the targeted biosyn-
thetic precursor (5) from the previously synthesized inter-
mediate 64 (Scheme 2). Thus, diol 6 was successfully
converted to hydroxy lactone 7 through the action of
p-TsOH (87% yield) and then deoxygenated through a
two-step sequence involving thiocarbonate formation
[PhOC=SCl] and reduction (n-Bu3SnH, AIBN)6 to afford
the expected β-keto lactone in 78% overall yield. NMR
spectroscopic analysis of the latter compound revealed its
exclusive existence in CDCl3 solution as its enol form 8
(Scheme 2). An X-ray crystallographic analysis of a single
crystal of this compound obtained from CH2Cl2/hexanes
(mp 163ꢀ166 °C, dec.) confirmed its enolic structure in the
solid state and proved the Z-geometry of its enol system
(see ORTEP, Scheme 2). Exposure of enol 8 to DDQ and
K2CO3 led to the corresponding R-hydroxy ketone, whose
dehydration with Martin’s sulfurane7 gaveketofuranone 9
in 73% overall yield.
(6) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans.
1975, 1574–1585.
(7) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327–
4329.
Org. Lett., Vol. 13, No. 20, 2011
5709

Scheme 4. Biomimetic Synthesis of Hirsutellones A (1), B (2),
and C (3) and 10,20,17-epi-Hirsutellone C (10,20,17-epi-3)
Scheme 5. Conversion of Epoxy Amide 11 to 10,20,17-epi-Hir-
sutellone C (10,20,17-epi-3)
isomer of the unsaturated hydroxy amide as confirmed later
in the sequence (see ring closure 13 f5þ 20-epi-5, Scheme 4).
Exposure of 12 to DMP then produced keto amide 14
which spontaneously cyclized, furnishing a mixture of 17,10-
dehydrohirsutellone B (5, major) and 20-epi-17,10-dehydro-
hirsutellone B (20-epi-5, minor), two labile intermediates
(not detectable by TLC) whose structures were inferred
from their successful conversion to the naturally occurring
hirsutellones A (1), B (2), and C (3), and 10,20,17-epi-
hirsutellone C (10,20,17-epi-3), respectively, as shown in
Scheme 4. Synthetic hirsutellones A, B, and C exhibited
identical physical properties to those reported for the natural
products.1 Exhibiting spectroscopic and spectrometric data
consistent with its structure, compound 10,20,17-epi-3 has
not been reported as a natural product as yet. However, it
should not be a surprise if it is found in nature in the future,
given the ease by which it is formed from 20-epi-5 (17,10-
dehydro-20-epi-hirsutellone B). 10,20,17-epi-Hirsutellone C
(10,20,17-epi-3) was also obtained from epoxy hydroxy amide
11 by DMP oxidation as shown in Scheme 5 (85% yield).
The described chemistry provides support for the Isaka
biosynthetic hypothesis for hirsutellones AꢀC (1ꢀ3) and
constitutes the first total synthesis of hirsutellones A (1), C
(3), and 10,20,17-epi-hirsutellone C (10,20,17-epi-3) as well as
a second synthesis of hirsutellone B (2).
With the procurement of keto furanone 9, only amida-
tion, oxidation, and ring closure remainedinorder toreach
the targeted molecule (5). The direct amidation of 9 proved
difficult, yielding only 17% of the desired hydroxy amide (12)
after 48 h of heating in neat NH3 (85% based on recovered 9)
as shown in Scheme 3. In contrast, epoxidation of 9 with
H2O2 under basic conditions (aq. NaOH) at 0 °C led to
epoxy lactone 10 in 90% yield within 30 min (single diaster-
eoisomer, R-stereochemistry assigned by NOE studies; see
Scheme 3). Pleasantly, ammonolysis of the lactone moiety of
epoxy lactone 10 proceeded smoothly to afford epoxy
hydroxy amide 11 in excellent yield (91%), requiring only
bubbling of NH3 gas through a solution of the substrate in
MeOH/THF/H2O (4:4:1) at 0 °C for 1 h. The latter com-
pound was then deoxygenated through the action of SmI2,
leading to the desired unsaturated hydroxy amide 12 in 67%
yield.8 In this reaction the initially formed 1,2-diol suffers
stereoselective dehydration to afford the desired geometrical
Acknowledgment. Within The Scripps Research Insti-
tute, we thank Drs. D.-H. Huang and L. Pasterneck for
NMR spectroscopic assistance, as well as Drs. G. Siuzdak
and R. Chadha for mass spectrometric and X-ray crystal-
lographic assistance, respectively. Financial support for this
work was provided by the National Institutes of Health
(USA) (R01 AI055475) and the Skaggs Institute for Re-
search, as well as fellowships from Bristol-Myers Squibb (to
D.S.) and the Natural Sciences and Engineering Research
Council, Canada (postdoctoral fellowship to T.R.W.). We
thank Dr. M. Isaka and Prof. Y. Thebtaranonth from the
National Center for Genetic Engineering and Biotech-
nology (Thailand) for samples of natural hirsutellones.
Supporting Information Available. Schemes with re-
spective reagents and conditions, experimental proce-
dures, characterization data for all compounds reported
and cif file (pdf). This material is available free of charge
via the Internet at http://pubs.acs.org.
(8) Molander, G. A.; Hahn, G. J. Org. Chem. 1986, 51, 2596–2599.
5710
Org. Lett., Vol. 13, No. 20, 2011