Cephalosporolide B serving as a versatile synthetic precursor: Asymmetric biomimetic total syntheses of cephalosporolides C, E, F, G, and (4-OMe-)G

Article abstract of DOI:10.1021/ol402913m

Cephalosporolide B (Ces-B) was efficiently synthesized and exploited for the first time as a versatile biomimetic synthetic precursor for the chemical syntheses of not only cephalosporolides C, G, and (4-OMe-) G via a challenging diastereoselective oxa-Michael addition but also the structurally unprecedented cephalosporolides E and F via a novel biomimetic ring-contraction rearrangement. These findings provide the first direct chemical evidence that Ces-B may be the true biosynthetic precursor of cephalosporolides.

Full text of DOI:10.1021/ol402913m

ORGANIC
LETTERS
2013
Vol. 15, No. 22
5850–5853
Cephalosporolide B Serving as a Versatile
Synthetic Precursor: Asymmetric
Biomimetic Total Syntheses of
Cephalosporolides C, E, F, G, and
(4-OMe-)G
Liyan Song, Yuan Liu, and Rongbiao Tong*
Department of Chemistry, The Hong Kong University of Science and Technology,
Clear Water Bay, Kowloon, Hong Kong, China
rtong@ust.hk
Received October 9, 2013
ABSTRACT
Cephalosporolide B (Ces-B) was efficiently synthesized and exploited for the first time as a versatile biomimetic synthetic precursor for the
chemical syntheses of not only cephalosporolides C, G, and (4-OMe-) G via a challenging diastereoselective oxa-Michael addition but also the
structurally unprecedented cephalosporolides E and F via a novel biomimetic ring-contraction rearrangement. These findings provide the first
direct chemical evidence that Ces-B may be the true biosynthetic precursor of cephalosporolides.
Biomimetic synthesis inspired by a biosynthesis (or
biosynthetic hypothesis) of natural products has been well
recognized as a highly efficient synthetic tactic for chemical
synthesis of complex molecules (e.g., natural products).1
It provides not only chemical evidence for its biogenesis
origin but also a new venue for synthesis of other structure-
related naturally occurring products. However, it is still a
formidable synthetic task to mimic an enzyme-mediated
transformation that involves a cascade reaction process or
rearrangement of structural skeletons. Another challenge
for biomimetic synthesis is to imitate the strategy used in
nature for construction of natural product collections from
a common intermediate or interconversions of natural
products within a family.2 This natural synthetic strategy
received less attention from synthetic communities due to
the structural diversity and scarcity of the parent natural
products. In this context, we herein report a strategy that led
to asymmetric biomimetic total syntheses of five natural
products from a parent natural product within the family,
which successfully mimicked two biosynthesis-inspired
processes: (i) a novel ring-contraction rearrangement of
10-membered lactones (cephalosporolides B, C, and G) to
5,5-spiroketal-cis-fused-γ-lactone (cephalosporolides E and
F) and (ii) a diastereoselective intermolecular oxa-Michael
addition (Scheme 1).
Cephalosporolides BꢀG (Figure 1) were isolated by
Hanson3 and co-workers from the industrial fermenta-
tion of the fungus Cephalosporium aphidicola, ACC 3490.
(1) For selected reviews on biomimetic synthesis of natural products,
see: (a) Poupon, E.; Nay, B. Biomimetic Organic Synthesis; Wiley-VCH:
Weinheim, 2011. (b) Bulger, P. G.; Bagal, S. K.; Marquez, R. Nat. Prod. Rep.
2008, 25, 254. (c) de la Torre, M. C.; Sierra, M. A. Angew. Chem., Int. Ed.
2004, 43, 160.
(2) For selected examples, see: (a) Flyer, A. N.; Si, C.; Myers, A. G.
Nat. Chem. 2010, 2, 886. (b) Jones, S. B.; Simmons, B.; Mastracchio, A.;
MacMillan, D. W. C. Nature 2011, 475, 183. (c) Wang, J.; Chen, S.-G.;
Sun, B.-F.; Lin, G.-Q.; Shang, Y.-J. Chem.;Eur. J. 2013, 19, 2539. (d)
Li, H.; Wang, X.; Hong, B.; Lei, X. J. Org. Chem. 2013, 78, 800.
r
10.1021/ol402913m
Published on Web 11/06/2013
2013 American Chemical Society

Scheme 1. Synthetic Plans and Hanson’s Biosynthetic
Hypothesis of Cephalosporolides E and F
Figure 1. Representative cephalosporolides.
The molecular structures of cephalosporolides were eluci-
dated by extensive NMR studies, and the relative config-
uration of Ces-C (2) and Ces-E (7) was unambiguously
determined by single-crystal X-ray diffraction. Interest-
ingly, some of these cephalosporolides were also isolated
recently from other natural sources such as Cordyceps
militaris BCC 2816,4 Beauveria bassiana,5 and/or wood
decay fungus Armillaria tabescens (strain JNB-OZ344).6
Structurally, these cephalosporolides (1ꢀ6) are 10-membered
lactones (namely decanolides) with a methyl group at C9,
resembling to other bioactive decanolides.7 However, the
structural skeleton of cephalosporolides E and F (Ces-E, 7,
and Ces-F, 8) characterized by the presence of 5,5-spiro-
ketal-cis-fused-γ-lactone was unprecedented at the time of
their isolation but found in other recently isolated natural
products such as cephalosporolides H and I, penisporo-
lides, and ascospiroketals.8
the synthetically challenging 10-membered lactone. How-
ever, it has not been reported that a unified synthetic stra-
tegy would lead to total syntheses of both 10-membered
cephalosporolides and the unique 5,5-spiroketal-cis-fused-
γ-lactones Ces-E and Ces-F. The combination of the wide
occurrence in nature and potential biological activity
coupled with the structural novelty and complexity of
Ces-E and Ces-F prompted us to develop a biomimetic
divergent synthetic strategy for the cephalosporolide
family, especially the potential transformations of the 10-
membered lactones to 5,5-spiroketal-cis-fused-γ-lactones.
Our synthetic strategy (Scheme 1) was primarily inspired
by Hanson’s biogenetic hypothesis3 of Ces-E and Ces-F,
which might arise from dehydrative ring contraction of
Ces-C via hydrolysis, lactonization, and acetalization.11
However, their attemptstothe chemicalconversion of Ces-
C into Ces-E and Ces-F in the laboratory were unsuccess-
ful. Intrigued by the employment of Ces-B for hypothetic
biosynthesis of tenuipyrone12 and pyridomacrolidin,13 we
envisioned that Ces-B could also be the biosynthetic pre-
cursor of cephalosporolides via a diastereoselective inter-
molecular oxa-Michael addition14 and/or Hanson’s ring-
contraction rearrangement (Ces-E and Ces-F, Scheme 1).
In addition, Ces-C and Ces-G, if available from Ces-B,
could be explored to verify Hanson’s biosynthetic hypoth-
esis under the conditions optimized for ring contraction
rearrangement of Ces-B.
Although the biological activity profiles of cephalospor-
olides have not been fully demonstrated, they have re-
ceived considerable synthetic attention9,10 partly because
of the novel structural skeleton of Ces-E and Ces-F and/or
(3) (a) Ackland, M. J.; Hanson, J. R.; Hitchcock, P. B.; Ratcliffe,
A. H. J. Chem. Soc., Perkin Trans. 1 1985, 843. (b) Farooq, A.; Gordon,
J.; Hanson, J. R.; Takahashi, J. A. Phytochemistry 1995, 38, 557.
(4) Rukachaisirikul, V.; Pramjit, S.; Pakawatchai, C.; Isaka, M.;
Supothina, S. J. Nat. Prod. 2004, 67, 1953.
ꢀ
(5) Oller-Lopez, J. L.; Iranzo, M.; Mormeneo, S.; Oliver, E.; Cuerva,
J. M.; Oltra, J. E. Org. Biomol. Chem. 2005, 3, 1172.
(6) Herath, H. M. T. B.; Jacob, M.; Wilson, A. D.; Abbas, H. K.;
Nanayakkara, N. P. D. Nat. Prod. Res. 2012, 1.
(10) (a) Fernandes, R. A.; Ingle, A. B. Synlett 2010, 158. (b) Brimble,
M. A.; Finch, O. C.; Heapy, A. M.; Fraser, J. D.; Furkert, D. P.;
O’Connor, P. D. Tetrahedron 2011, 67, 995. (c) Ramana, C. V.;
Suryawanshi, S. B.; Gonnade, R. G. J. Org. Chem. 2009, 74, 2842. (d)
Tlais, S. F.; Dudley, G. B. Beilstein J. Org. Chem. 2012, 8, 1287. (e)
Chang, S.; Britton, R. Org. Lett. 2012, 14, 5844.
(11) Oltra observed the facile conversion of coisolated bassianolone
to Ces-E and Ces-F via acetalization and proposed that bassianolone
might be the true biosynthetic precursor. See refs 5 and 25.
(12) (a) Asai, T.; Chung, Y.-M.; Sakurai, H.; Ozeki, T.; Chang, F.-R.;
Yamashita, K.; Oshima, Y. Org. Lett. 2012, 14, 513. (b) Song, L.; Yao,
H.; Zhu, L.; Tong, R. Org. Lett. 2013, 15, 6.
€
(7) For reviews, see: (a) Drager, G.; Kirschning, A.; Thiericke,
R.; Zerlin, M. Nat. Prod. Rep. 1996, 13, 365. (b) Ferraz, H. M. C.;
Bombonato, F. I.; Longo, L. S., Jr. Synthesis 2007, 3261. (c) Riatto,
V. B.; Pilli, R. A.; Victor, M. M. Tetrahedron 2008, 64, 2279. For selected
syntheses of other decanolides, see: (d) Vadhadiya, P. M.; Puranik,
V. G.; Ramana, C. V. J. Org. Chem. 2012, 77, 2169. (e) Ramana, C. V.;
Khaladkar, T. P.; Chatterjee, S.; Gurjar, M. K. J. Org. Chem. 2008, 73,
3817. (f) Kobayashi, Y.; Asano, M.; Yoshida, S.; Takeuchi, A. Org. Lett.
2005, 7, 1533.
(8) (a) Li, X.; Yao, Y.; Zheng, Y.; Sattler, I.; Lin, W. Arch. Pharm.
Res. 2007, 30, 812. (b) Li, X.; Sattler, I.; Lin, W. J. Antibiot. 2007, 60, 191.
€
(c) Seibert, S. F.; Krick, A.; Eguereva, E.; Kehraus, S.; Konig, G. M.
Org. Lett. 2007, 9, 239.
(9) (a) Krishna, P. R.; Sreeshailam, A. Synlett 2008, 2795. (b)
Barradas, S.; Urbano, A.; Carreno, M. C. Chem.;Eur. J. 2009, 15,
9286. (c) Ma, B.; Zhong, Z.; Hu, H.; Li, H.; Zhao, C.; Xie, X.; She, X.
Chem.;Asian J. 2013, 8, 1391.
(13) Baldwin, J. E.; Adlington, R. M.; Conte, A.; Irlapati, N. R.;
Marquez, R.; Pritchard, G. J. Org. Lett. 2002, 4, 2125.
€
(14) For recent reviews, see: (a) Nising, C. F.; Brase, S. Chem. Soc.
~
€
Rev. 2008, 37, 1218. (b) Nising, C. F.; Brase, S. Chem. Soc. Rev. 2012, 41,
988.
Org. Lett., Vol. 15, No. 22, 2013
5851

Scheme 2. Total Synthesis of Cephalosporolide B (1)a
Scheme 3. Total Syntheses of Cephalosporolides C and G and
4-OMe-cephalosporolide G from Cephalosporolide B
a Abbreviations: TBSCl, chloro-tert-butyldimethylsilane; TBAF, tet-
rabutylammonium floride; DEAD, diethyl azodicarboxylate; m-CPBA,
meta-chloroperoxybenzoic acid; PCC, pyridinium chlorochromate.
To achieve a practical total synthesis of the key Ces-B
with sufficient quantity for subsequent biomimetic synthetic
studies, we chose oxidative ring expansion of β-hydro-
xyethers developed by Ferraz15 as the key step to construct
the 10-membered lactone (Scheme 2). The enantiomerically
pure rhododendrol (11)16 was chemoselectively protected as
tert-butyldimethylsilyl ether 12. Phenol dearomatization17
of 12 with PhI(OAc)2, desilylation, and subsequent TsOH-
promoted oxa-Michael cyclization provided bicyclic
ether 13 as a single diastereomer.18 Luche reduction of 13,
chemoselective silylation of the secondary alcohol, and
hydroxyl-directed epoxidation with m-CPBA provided ep-
oxide 14, which upon treatment of PCC underwent oxida-
tive ring expansion to give the 10-membered lactone 15.9b
Rh-catalyzed deoxygenation19 of epoxide 15 with dimethyl
diazomalonate unmasked the cis-alkene to furnish the
(þ)-Ces-B (1)20 after desilylation with HF-pyridine com-
plex. Noteworthy was the employment of epoxide as an
unusual protecting group of cis-alkene to avoid oxidative
rearrangement of tertiary alcohol in the PCC oxidation. All
spectroscopic data of our synthetic Ces-B were in good
agreement with those reported in the literature.3a,9c
stereoselectivity, and reversibility issues,14 we were de-
lighted to know that She9c and Xie documented a success-
ful oxa-Michael addition of MeOH to 16, providing
(þ)-4-OMe-Ces-G (5). Analogously, we found that cam-
phorsulfonic acid (CSA) or Amberlyst-15 (A-15) effec-
tively promoted the syn-Michael addition of MeOH to
Ces-B, affording (þ)-4-OMe-Ces-G (5) in 72% yield as a
single diastereomer. Encouraged by this result, we em-
ployed benzyl alcohol for the similar syn-oxa-Michael
addition to Ces-B. Not surprisingly, after Pd-catalyzed
hydrogenative debenzylation of 4-OBn-Ces-G (17),
(þ)-Ces-G (4)9b could be obtained in 57.3% yield over
two steps.20 Note: although enzyme-catalyzed Michael
addition of water to conjugated carbonyl compounds is
well-known,21 the corresponding nonenzymatic process
remains very limited.22 These exciting findings drove us
to explore the possibility of an anti-oxa-Michael addition
of benzyl alcohol to Ces-B, which after debenzylation
would be expected to give Ces-C (2), a biosynthetic
precursor of Ces-E and Ces-F in Hanson’s hypothesis.
Unfortunately, in contrast to She’s observation,9c we were
not able to achieve such anti-oxa-Michael addition under
various basic conditions, which only led to decomposition of
Ces-B. Inspired by Evans’ example of cascade acetalization/
oxa-Michael cyclization,23 we found that the similar cas-
cade reaction of aromatic aldehyde and Ces-B could
proceed smoothly to provide a diastereomeric mixture
of bicyclic acetal 19 in good yield. Removal of the aryl
acetal by CAN oxidation in a buffered solution produced
the expected (þ)-Ces-C (2) in excellent yield as a single
With Ces-B (1, ∼200 mg) in hand, we set out to exploit
it as a biomimetic synthetic precursor for Ces-C, Ces-G,
and 4-OMe-Ces-G via diastereoselective intermolecular
oxa-Michael addition (Scheme 3). Although it is well
recognized that intermolecular oxa-Michael addition has
suffered from many drawbacks such as low reactivity, low
(15) (a) Ferraz, H. M. C.; Longo, L. S., Jr. Org. Lett. 2003, 5, 1337. (b)
Ferraz, H. M. C.; Longo, L. S., Jr. J. Org. Chem. 2007, 72, 2945.
(16) Yuasa, Y.; Shibuya, S.; Yuasa, Y. Synth. Commun. 2003, 33,
1469.
(17) (a) For a representative review, see: Roche, S. P.; Porco, J. A., Jr.
Angew. Chem., Int. Ed. 2011, 50, 4068. For selected recent examples, see:
(b) Zhao, K.; Cheng, G.-J.; Yang, H.; Shang, H.; Zhang, X.; Wu, Y.-D.;
Tang, Y. Org. Lett. 2012, 14, 4878. (c) Brown, P. D.; Willis, A. C.;
Sherburn, M. S.; Lawrence, A. L. Org. Lett. 2012, 14, 4537.
(18) Note: 13 could be synthesized from 11 in two steps with
comparable overall yield on small scale via oxone-mediated dearoma-
tization and TsOH-promoted oxa-Michael cyclization (ref 9b); however,
dearomatization on large scale (>18 mmol) was problematic. See
ref 17b,17c.
(21) (a) Jin, J.; Oskam, P. C.; Karmee, S. K.; Straathof, A. J. J.;
Hanefeld, U. Chem. Commun. 2010, 46, 8588. (b) Gawron, O.; Fondy,
T. P. J. Am. Chem. Soc. 1959, 81, 6333. (c) Willadsen, P.; Eggerer, H.
Eur. J. Biochem. 1975, 54, 247. (d) Agnihotri, G.; Liu, H. Bioorg. Med.
Chem. 2003, 11, 9. (e) Mohrig, J. R.; Moerke, K. A.; Cloutier, D. L.;
Lane, B. D.; Person, E. C.; Onasch, T. B. Science 1995, 269, 527.
ꢁ
(22) Boersma, A. J.; Coquiere, D.; Geerdink, D.; Rosati, F.; Feringa,
B. L.; Roelfes, G. Nat. Chem 2010, 2, 991.
(19) Martin, M. G.; Ganem, B. Tetrahedron Lett. 1984, 25, 251.
(20) See the Supporting Information.
(23) Evans, D. A.; Gauchet-Prunet, J. A. J. Org. Chem. 1993, 58,
2446.
5852
Org. Lett., Vol. 15, No. 22, 2013

Table 1. Ring-Contraction Rearrangement of Cephalosporolide
B to Cephalosporolides E and Fa
Scheme 4. Realization of Hanson’s Hypothesis and Proposed
Mechanism of Dehydrative Ring-Contraction Rearrangement
to Cephalosporolides E and F
entry acid (equiv) solvents (ratio) time (h) yield (%, 7/8)
1
2
3
4
5
6
7
HClb
THF
1
12
12
12
8
<10
TsOH (5)
A-15 (10)
THF/H2O (9/1)
THF/H2O (9/1)
NR
NR
TFA/THF/H2O (1/1/1)
TFA/THF/H2O (3/1/1)
TFA/THF/H2O (9/1/1)
TFA/CH2Cl2/H2O (3/1/1)
40 (3:1)c
79 (3:1)
complex
complex
2
2
a Reaction with Ces-B (20 mg, 0.1 mmol) was run at 0.05 M at rt;
isolated yield. b One drop (∼5 mg) of concd HCl was added to 2.0 mL of
THF solution at 0 °C. c Recovery of 42% yield of Ces-B. NR: no reaction.
diastereomer. This constitutes the first, asymmetric total
synthesis of Ces-C. Alternatively, Ces-C could be synthe-
sized from epoxidation of Ces-B followed by SmI2-
mediated reductive epoxide ring-opening24 of 20. Our
synthetic (þ)-Ces-C was fully confirmed by extensive
NMR studies and single-crystal X-ray diffraction. How-
ever, we noticed that the NMR data of our synthetic Ces-C
were not in well agreement with those reported by Hanson,
but with X-ray diffraction analysis of natural and synthetic
samples we believed the NMR data for Ces-C might be
erroneously reported.25
Finally, we set out to verify our key hypothesis that
Ces-B could be the direct synthetic precursor of Ces-E
and -F (Scheme 1). Because of decomposition under basic
conditions, we focused on the ring-contraction rearrange-
ment of Ces-B under acidic conditions (Table 1). To our
delight, we observed for the first time that Ces-E and Ces-F
could be generated from Ces-B upon addition of one
drop of concentrated HCl to the THF solution of Ces-B
(entry 1). Further optimization (entries 2ꢀ7) led us to
identify the trifluoroacetic acid (TFA) as the best acid for
the ring-contraction rearrangement of Ces-B (1), affording
a 3:1 mixture of Ces-E and Ces-F in 79% combined yield
(entry 5), favoring the thermodynamically more stable
Ces-E. This is the first example that demonstrated a ring-
contraction rearrangement of a ten-membered lactone to
5,5-spiroketal-cis-fused-γ-lactone, a cascade process mim-
icking Hanson’s hypothesis. This exciting discovery
prompted us to further examine Hanson’s hypothesis:
dehydrative ring contraction of Ces-C into Ces-E and
Ces-F (Scheme 4). In sharp contrast to Hanson’s results,
we foundthatunder our optimized condition Ces-C under-
went efficient dehydrative ring contraction to provide a
3:1 mixture of Ces-E and Ces-F in excellent yield. Most
strikingly, Ces-G, a diastereomer of Ces-C, was able to
rearrange to afford a mixture of Ces-E and Ces-F in
comparable yield with the same diastereomeric ratio. The
inversion mechanism of C4 stereogenic center of Ces-G
remains unknown, however, the inconsequence of stereo-
genic center at C4 in the course of rearrangement (Ces-C
and Ces-G) might suggest that Ces-B might be the true
biosynthetic precursor of Ces-E and Ces-F. Taking all
these evidence together, we proposed that Ces-E and
Ces-F might arise from Ces-B through hemiacetal forma-
tion (21), macrolactone opening to carboxylic acid 23,
γ-lactone formation (24) by SN20 substitution and acid-
promoted spiroketalization. This hypothesis also ex-
plained the coisolated furan 22, which may be generated
from dehydrative aromatization of 23.
In conclusion, we have achieved a practical total synth-
esis of cephalosporolide B, which has been successfully
exploited as a versatile synthetic precursor for biomimetic
total syntheses of cephalosporolides C, E, F, G, and
(4-OMe-)G via an oxa-Michael addition and/or a novel
biomimetic ring-contraction rearrangement. These studies
suggested that cephalosporolide B might be the true
biogenetic precursor of other cephalosporolides, which
may imply the biogenetic relationship of the cephalospor-
olide family and find applications in total syntheses of
other bioactive natural products. In addition, the 28-year
Hanson’s biogenetic hypothesis of cephalosporolides E
and F was synthetically verified for the first time in the
laboratory.
Acknowledgment. This research was financially sup-
ported by startup funding (R9309) from HKUST and
Research Grant Council of Hong Kong (ECS 605912
and DAG12CS03). We are also grateful to Prof. Dr. Ian
Williams (HKUST) for the single-crystal X-Ray diffrac-
tion analysis.
Supporting Information Available. Detailed experimen-
tal procedures, characterization, and copies of 1H and 13
C
(24) Molander, G. A.; Hahn, G. J. Org. Chem. 1986, 51, 2596.
(25) This phenomenon is rare in the literature. Surprisingly, we
observed that our NMR spectra for Ces-C were identical to those of
Ces-J and bassianolone, and we suspected that Ces-C was the correct
structure for Ces-J and bassianolone. For more details, see the Support-
ing Information.
NMR spectra of new compounds. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 22, 2013
5853