Protecting-Group-Free Enantioselective Synthesis of (-)-Pallavicinin and (+)-Neopallavicinin

Article abstract of DOI:10.1002/anie.201506575

The first enantioselective synthesis of (-)-pallavicinin and (+)-neopallavicinin has been achieved in 15 steps. The described synthesis avoids protecting-group manipulations by synthesis designs predicated on highly chemo- and stereoselective transformations. Highlights of the synthesis include a palladium-catalyzed enantioselective decarboxylative allylation to form the chiral all-carbon quaternary stereocenter, a palladium-catalyzed oxidative cyclization to assemble the [3.2.1]-bicyclic moiety, and an unprecedented LiBHEt₃-induced fragmentation/protonation of an α-hydroxy epoxide to form the α-furan ketone with the desired configuration.

Full text of DOI:10.1002/anie.201506575

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201506575
German Edition: DOI: 10.1002/ange.201506575
Natural Products
Protecting-Group-Free Enantioselective Synthesis of (À)-Pallavicinin
and (+)-Neopallavicinin
Bin Huang, Lei Guo, and Yanxing Jia*
Abstract: The first enantioselective synthesis of (À)-pallavici-
nin and (+)-neopallavicinin has been achieved in 15 steps. The
described synthesis avoids protecting-group manipulations by
synthesis designs predicated on highly chemo- and stereose-
lective transformations. Highlights of the synthesis include
a palladium-catalyzed enantioselective decarboxylative allyla-
tion to form the chiral all-carbon quaternary stereocenter,
a palladium-catalyzed oxidative cyclization to assemble the
[3.2.1]-bicyclic moiety, and an unprecedented LiBHEt3-
induced fragmentation/protonation of an a-hydroxy epoxide
to form the a-furan ketone with the desired configuration.
centers, including an all-carbon quaternary center, and
a bridge ring of the left-part bicyclo[3.2.1] moiety. The
unique structural features and potential bioactivities of
(+)-1 and (À)-2 make them attractive targets for total
synthesis.[5,6]
In this context, the group of Wong reported the first, and
thus far, the only total synthesis of (Æ)-1 and (Æ)-2 in
a biomimetic manner from the (Æ)-Wieland-Miescher ketone
in 2006.[5a]
Despite tremendous advances of new methodologies and
strategies in organic chemistry, protecting-group-free (PGF)
total syntheses of complex natural products still present
a highly challenging task.[7,8] As a continuation of our ongoing
projects focused on improving the efficiency of complex
natural products synthesis,[9] herein we report the first
asymmetric total synthesis of (À)-1 and (+)-2 without the
use of protecting groups. Moreover, an unprecedented
LiBHEt3 (super hydride)-induced fragmentation/protonation
of an a-hydroxy epoxide was discovered.
Our retrosynthetic analysis of (À)-1 and (++)-2 is illus-
trated in Scheme 1. We envisioned that both (À)-1 and (+)-2
could be generated from the lactone 3 by oxa-Michael
addition and subsequent aldol condensation. The compound
3 could then be derived from the ketone 4 by nucleophilic
methylation of the C8 carbonyl with subsequent peroxy acid
oxidation of the furan moiety. The expeditious assembly of 4
with the desired a-C9 configuration of the furan substituent
could be derived from the a-hydroxy epoxide 5 by an
unprecedented LiBHEt3-induced fragmentation/protonation,
D
iterpenoids from liverworts often have interesting biolog-
ical activities such as antifungal, antimicrobial, cytotoxic,
insect antifeedant, insecticidal, and muscle relaxants.[1]
(+)-Pallavicinin [(+)-1] and (À)-neopallavicinin [(À)-2],
two structurally complex secolabdane-type diterpenoids,
were first isolated by Wu and co-workers from the Taiwanese
liverwort Pallavicinia subciliata, and then by Lou and co-
workers from the Chinese liverwort Pallavicinia ambigua
(Figure 1).[2,3] (+)-Pallavicinin (1), along with other biosyn-
Figure 1. Structures of (À)-pallavicinin (1) and (+)-neopallavicinin (2).
thesis-relative natural products had also been identified from
the Japanese liverwort Pallavicinia subciliata by Asakawa and
co-workers.[4] The absolute configurations of (+)-1 and (À)-2
were unambiguously assigned by single-crystal X-ray diffrac-
tion and circular dichroism (CD) analyses.[3] Their unique
structures feature a novel ladder-shaped/cagelike [6-5-5-5]
tetracyclic skeleton bearing seven contiguous stereogenic
[*] B. Huang, L. Guo, Prof. Y. Jia
State Key Laboratory of Natural and Biomimetic Drugs, School of
Pharmaceutical Sciences, Peking University
Xue Yuan Rd. 38, Beijing 100191 (China)
E-mail: yxjia@bjmu.edu.cn
Prof. Y. Jia
State Key Laboratory of Bioorganic and Natural Products Chemistry,
Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, Shanghai 200032 (China)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201506575.
Scheme 1. Retrosynthetic analysis of (À)-1 and (+)-2. TMS=trimethyl-
silyl.
Angew. Chem. Int. Ed. 2015, 54, 13599 –13603
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13599

.
Angewandte
Communications
which was serendipitously discovered during attempts to
utilize the LiBHEt3-mediated Payne rearrangement of 5.[10] It
was very difficult to install the furan with the a configuration
at the C9-position of 4 because the furan moiety is located on
the endo-face of this concave skeleton, and nucleophilic
attack or electrophilic attack would preferentially occur on
the less sterically hindered exo-face.[6a,11] The epoxide 5 could
be accessed from the [3.2.1]bicyclic ketone 6 by functional-
the nucleophile (6a in Scheme 2).[16] Double methylation of
10 produced the desired diene intermediate 11 in 93% yield.
Selective allylic oxidation at C9 of 11 with an excess of
tBuO2H (TBHP) in the presence of a catalytic amount of
SeO2 afforded the corresponding allylic alcohol 12 as the sole
stereoisomer (Scheme 2).[17] The hydroxy-directed epoxida-
tion of 12 with VO(acac)2/TBHP furnished the expected
epoxide 13 as the only product.[18] The subsequent oxidation
of 13 with Dess–Martin periodinane (DMP) provided the
diketone 14 in 87% yield, and its structure was confirmed by
the X-ray crystallography analysis.[19] Importantly, the opti-
cally pure diketone 14 (> 99% ee) could be readily obtained
after one recrystallization, thus providing a basis for the
enantioselective syntheses of (À)-1 and (+)-2. Chemo- and
stereoselective addition of trimethylsilyl furan lithium to the
C9 carbonyl group of 14 delivered the desired epoxide 5 in
72% yield.[20,21]
À
group manipulations. Accordingly, the C2 C8 bond of 6 could
be assembled by a palladium-catalyzed oxidative cyclization
of the known chiral cyclohexenone 7,[12,13] which has been
prepared by a palladium-catalyzed enantioselective decar-
boxylative allylation of 8, elegantly developed by Stoltz and
co-workers.[14,15]
Our synthesis commenced with the preparation of the
chiral cyclohexenone 7 in 85% ee following Stoltzꢀs proce-
dure, thus constructing the all-carbon quaternary stereogenic
center (Scheme 2).[14a] Treatment of 7 with LDA and TBSCl
With 5 in hand, LiBHEt3-mediated Payne rearrangement
was initially tested (Scheme 3). Surprisingly, treatment of 5
Scheme 3. Unexpected LiBHEt3-induced fragmentation of 5.
with LiBHEt3 at 308C not only afforded the epoxide
migration product 15 in 35% yield, but also led to the
isolation of an unexpected ketone (4) in 45% yield, and it is
the key late-stage intermediate in our retrosynthetic design.
Further optimizing the reaction conditions showed that 4
could be obtained as the sole product in 84% yield by
increasing the reaction temperature to 608C. In contrast,
decreasing the reaction temperature to 08C only gave the
epoxide 15 in 90% yield. To the best of our knowledge, this
LiBHEt3-induced fragmentation of such a-hydroxy epoxide
to the ketone has not been reported previously.[22]
To have preliminary insight on such an unprecedented
fragmentation reaction, the epoxide 15, resulting from the
Payne rearrangement was subjected to the reaction condi-
tions (LiBHEt3 at 308C or 608C), but it did not react to give
the desired ketone 4. This result implied that the rearrange-
ment product 15 was not the exact precursor of the
fragmentation reaction. Based on the literature and the
above facts, a plausible mechanism for the fragmentation/
protonation cascade process is proposed in Scheme 4. Firstly,
the ketone moiety on C3 is reduced and the C9 hydroxy group
is deprotonated with LiBHEt3 to form the alkoxide anion A.
When the reaction is run at low temperature, the epoxide
Scheme 2. Synthesis of 5. LDA=lithium diisopropylamide, TBS=tert-
butyldimethylsilyl, DME=ethylene glycol dimethyl ether, DMP=Dess–
Martin periodinane, DMSO=dimethyl sulfoxide, HMPA=hexamethyl-
phosphoric triamide, THF=tetrahydrofuran, VO(acac)2 =vanadyl ace-
tylacetonate.
led to the corresponding TBS enol ether, which underwent
oxidative cyclization upon exposure to a catalytic amount of
Pd(OAc)2 under an atmosphere of O2 to give the desired
[3.2.1]bicyclic ketone 6 in 77% overall yield.[14] Conjugate
addition of a vinyl group to 6 afforded the ketone 10 with
a satisfactory diastereomeric ratio (d.r. = 10:1). The observed
high preference for re-face attack can be easily explained by
the steric hindrance between the allylic hydrogen atom and
13600 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 13599 –13603

Angewandte
Chemie
Scheme 4. Proposed mechanism for reaction of 5 with LiBHEt3.
migration is under kinetic control, and Payne rearrangement
occurrs to form 15 (path a). When the reaction is run at higher
temperature, it is under thermodynamic control and the
attack of the alkoxide anion at the less-substituted carbon
atom of the epoxide formed the oxetane intermediate C,[23]
which could readily undergo a fragmentation reaction to
generate the enolate D. Finally, D is quenched with a proton
from the less-hindered equatorial face (exo-face) of the
concave skeleton to produce 4.
Having set the a configuration of the furan group at C9 of
4, the completion of the syntheses of (À)-1 and (+)-2 is shown
in Scheme 5. At this stage, attempts on direct methylation of 4
resulted in the decomposition of the starting material. Dess–
Martin oxidation of 4 provided the diketone 16. To pursue
a protecting-group-free route, a challenging chemoselective
monomethylation at the C8 carbonyl group of 16 was then
examined. Initially methylation using 1.5 equivalents of MeLi
in THF yielded the desired product 17 in 40% yield,
accompanied by the undesired byproduct 18 in 12% yield,
as well as the recovery of starting material 16 in 28% yield.[24]
However, 17 and 16 could not be separated by column
chromatography. An assortment of reaction conditions with
various methylation reagents, solvents, additives, and temper-
atures were screened (see the Supporting Information), and it
was found that reaction of 4 with 5.0 equivalents of MeMgBr
in THF at À408C (two cycles) afforded the desired 17 in 80%
yield.
Scheme 5. Completing the total synthesis of (À)-1 and (+)-2.
m-CPBA=meta-chloroperoxybenzoic acid, DBU=1,8-diazabicyclo-
[5.4.0]undec-7-ene, DMAP=4-dimethylaminopyridine, Ms=methane-
sulfonyl.
(Scheme 6).[19] Further control experiments showed that 3b
could also be completely converted into the products 20 and
21 in 83% overall yield with the same ratio (20/21 = 2:1) by
prolonging the reaction time under the aforementioned basic
conditions. These results indicated that the undetected
intermediate 3a, which could be formed in situ from a fast
equilibration with 3b in the presence of DBU, quickly
underwent a subsequent intramolecular oxa-Michael addition
to yield the stereodefined compound 20.
Epoxidation of 17 with m-CPBA gave the unstable b,g-
butenolide 19, which was subsequently treated with DBU and
underwent a base-induced double-bond migration followed
by intramolecular oxa-Michael addition to form the key
tetracyclic compounds 20 and 21 in 30% and 15% yield,
respectively (Scheme 5). Notably, the key intermediate a,b-
butenolide 3b, rather than its C11 epimer 3a, could be readily
isolated in 48% yield after column chromatography, when the
reaction was quenched after 5 minutes, and its structure was
unequivocally confirmed by X-ray crystallographic analysis
Scheme 6. Conversion of 19 into 20 and 21.
Angew. Chem. Int. Ed. 2015, 54, 13599 –13603
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 13601

.
Angewandte
Communications
[9] a) H. Qin, Z. Xu, Y. Cui, Y. Jia, Angew. Chem. Int. Ed. 2011, 50,
4447; Angew. Chem. 2011, 123, 4539; b) W. Hu, H. Qin, Y. Cui, Y.
Jia, Chem. Eur. J. 2013, 19, 3139; c) L. Guo, F. Zhang, W. Hu, L.
Li, Y. Jia, Chem. Commun. 2014, 50, 3299.
[10] a) M. Dobler, J. C. Anderson, M. Juch, H.-J. Borschberg, Helv.
Chim. Acta 1995, 78, 292; b) R. M. Hanson, in Organic
Reactions, Vol. 60 (Eds.: L. E. Overman, et al.), Wiley, New
York, 2002; pp. 1 – 156.
[11] For example, bromination of the compound A gave the a-
bromide ketone B, and installation of the 1,4-butanolide moiety
to A afforded the product C. The addition of trimethylsilyl furan
lithium to the C9 carbonyl group of 14 delivered the epoxide 5
(Scheme 2).
Finally, the aldol reaction of 20 with acetaldehyde
followed by treatment of the corresponding adduct with
MsCl and Et3N provided the target molecule (À)-1 in 55%
over two steps, and its absolute configuration was confirmed
by X-ray crystallography (Scheme 5).[19] In the same manner,
21 can be transformed into the target molecule (+)-2, albeit
only in 12% yield. The physical data of our synthesized
products (À)-1 and (+)-2 are identical to those reported in the
literature (see the Supporting Information).[2,3,5a]
In summary, we have accomplished the first asymmetric
total synthesis of (À)-1 and (++)-2 in 15 steps from the known
cyclohexenone 7 without the use of protecting groups. The
success of this PGF synthesis was mainly dependent on
several highly chemo- and stereoselective reactions. The
present synthesis features a palladium-catalyzed enantiose-
lective decarboxylative allylation to form the chiral all-carbon
quaternary center, a palladium-catalyzed oxidative cycliza-
tion to assemble the [3.2.1]bicyclic moiety, and an unprece-
dented LiBHEt3-induced fragmentation/protonation of an a-
hydroxy epoxide to form the a-furan ketone with the desired
configuration.
[12] For seminal reports on stoichiometric palladium(II)-mediated
silyl enol ether oxidative cyclizations, see: a) A. S. Kende, B.
Roth, P. J. Sanfilippo, J. Am. Chem. Soc. 1982, 104, 1784; b) A. S.
Kende, B. Roth, P. J. Sanfilippo, T. J. Blacklock, J. Am. Chem.
Soc. 1982, 104, 5808.
[13] For the development of palladium(II)-catalyzed oxidative cyc-
lizations, see: a) M. Toyota, T. Wada, K. Fukumoto, M. Ihara, J.
Am. Chem. Soc. 1998, 120, 4916; b) M. Toyota, M. Rudyanto, M.
Ihara, J. Org. Chem. 2002, 67, 3374; For a review, see: c) M.
Toyota, M. Ihara, Synlett 2002, 1211; For selective synthetic
application, see: d) J. T. S. Yeoman, V. M. Mak, S. E. Reisman, J.
Am. Chem. Soc. 2013, 135, 11764; e) O. F. Jeker, E. M. Carreira,
Angew. Chem. Int. Ed. 2012, 51, 3474; Angew. Chem. 2012, 124,
3531; f) K. C. Nicolaou, G. S. Tria, D. J. Edmonds, M. Kar, J. Am.
Chem. Soc. 2009, 131, 15909; g) G. N. Varseev, M. E. Maier,
Angew. Chem. Int. Ed. 2009, 48, 3685; Angew. Chem. 2009, 121,
3739; h) M. Toyota, T. Asano, M. Ihara, Org. Lett. 2005, 7, 3929;
i) M. Toyota, T. Odashima, T. Wada, M. Ihara, J. Am. Chem. Soc.
2000, 122, 9036.
Acknowledgements
We are grateful to the National Natural Science Foundation
of China (Nos. 21290183, and 21572008) for their financial
support.
Keywords: epoxidation · natural products · palladium ·
terpenoids · total synthesis
How to cite: Angew. Chem. Int. Ed. 2015, 54, 13599–13603
Angew. Chem. 2015, 127, 13803–13807
[14] a) S. R. Levine, M. R. Krout, B. M. Stoltz, Org. Lett. 2009, 11,
289; b) J. T. Mohr, D. C. Behenna, A. M. Harned, B. M. Stoltz,
Angew. Chem. Int. Ed. 2005, 44, 6924; Angew. Chem. 2005, 117,
7084; c) D. C. Behenna, B. M. Stoltz, J. Am. Chem. Soc. 2004,
126, 15044; d) D. C. Behenna, J. T. Mohr, N. H. Sherden, S. C.
Marinescu, A. M. Harned, K. Tani, M. Seto, S. Ma, Z. Novµk,
M. R. Krout, R. M. McFadden, J. L. Roizen, J. A. Enquist Jr.,
D. E. White, S. R. Levine, K. V. Petrova, A. Iwashita, S. C. Virgil,
B. M. Stoltz, Chem. Eur. J. 2011, 17, 14199; For a review, see:
e) A. Y. Hong, B. M. Stoltz, Eur. J. Org. Chem. 2013, 2745. For
selective recent examples on application of the palladium(II)-
catalyzed enantioselective decarboxylative allylation in natural
product syntheses, see: f) Z. Li, S. Zhang, S. Wu, X. Shen, L. Zou,
F. Wang, X. Li, F. Peng, H. Zhang, Z. Shao, Angew. Chem. Int.
Ed. 2013, 52, 4117; Angew. Chem. 2013, 125, 4211; g) Z. Xu, Q.
Wang, J. Zhu, J. Am. Chem. Soc. 2013, 135, 19127; h) Z. Xu, Q.
Wang, J. Zhu, J. Am. Chem. Soc. 2015, 137, 6712.
[15] B. M. Trost, R. N. Bream, J. Xu, Angew. Chem. Int. Ed. 2006, 45,
3109; Angew. Chem. 2006, 118, 3181.
[16] M. Toyota, T. Wada, Y. Nishikawa, K. Yanai, K. Fukumoto, C.
Kabuto, Tetrahedron 1995, 51, 6927.
[17] a) M. A. Umbreit, K. B. Sharpless, J. Am. Chem. Soc. 1977, 99,
5526; b) S. C. Dolan, D. W. Holdup, M. Hutchison, J. MacMillan,
J. Chem. Soc. Perkin Trans. 1 1985, 651.
[18] a) K. B. Sharpless, R. C. Michaelson, J. Am. Chem. Soc. 1973, 95,
6136; b) E. D. Mihelich, K. Daniels, D. J. Eickhoff, J. Am. Chem.
[1] a) Y. Asakawa, Phytochemistry 2004, 65, 623; b) Y. Asakawa,
Phytochemistry 2001, 56, 297.
[2] a) C.-L. Wu, H.-J. Liu, H.-L. Uang, Phytochemistry 1994, 35, 822;
b) H.-J. Liu, C.-L. Wu, J. Asian Nat. Prod. Res. 1999, 1, 177.
[3] Z.-J. Li, H.-X. Lou, W.-T. Yu, P. -H. Fan, D.-M. Ren, B. Ma, M. Ji,
Helv. Chim. Acta 2005, 88, 2637.
[4] M. Toyota, T. Saito, Y. Asakawa, Chem. Pharm. Bull. 1998, 46,
178.
[5] a) X.-S. Peng, H. N. C. Wong, Chem. Asian J. 2006, 1, 111; For
total synthesis of other origin-relative natural products, see:
b) X.-S. Xu, Z.-W. Li, Y.-J. Zhang, X.-S. Peng, H. N. C. Wong,
Chem. Commun. 2012, 48, 8517; c) J.-Q. Dong, H. N. C. Wong,
Angew. Chem. Int. Ed. 2009, 48, 2351; Angew. Chem. 2009, 121,
2387.
[6] For synthetic studies of pallavicinin and neopallavicinin, see:
a) A. R. Hardin-Narayan, PhD thesis, University of California,
ˇ
ˇ
ˇ
Berkeley, 2011; b) M. Markovic, M. Duranovµ, P. KoóS, P.
Szolcsµnyi, T. Gracza, Tetrahedron 2013, 69, 4185.
[7] For recent reviews on the protecting-group-free syntheses, see:
a) I. S. Young, P. S. Baran, Nat. Chem. 2009, 1, 193; b) R. N.
Saicic, Tetrahedron 2014, 70, 8183; c) R. W. Hoffmann, Synthesis
2006, 3531.
[8] For selective recent examples on protecting-group-free synthe-
ses of complex natural products, see: a) S. Krüger, T. Gaich,
Angew. Chem. Int. Ed. 2015, 54, 315; Angew. Chem. 2015, 127,
320; b) Y.-M. Zhao, T. J. Maimone, Angew. Chem. Int. Ed. 2015,
54, 1223; Angew. Chem. 2015, 127, 1239.
13602 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 13599 –13603

Angewandte
Chemie
Soc. 1981, 103, 7690; c) B. M. Trost, L. Dong, G. M. Schroeder, J.
Am. Chem. Soc. 2005, 127, 10259.
[22] Literature search showed that two examples were reported
wherein ketones were obtained as side-products from a-hydroxy
epoxides by using either Yb(OTf)3 or [Cp2TiCl]. See: a) B.
Hauer, J. F. Bickley, J. Massue, P. C. A. Pena, S. M. Roberts, J.
Skidmore, Can. J. Chem. 2002, 80, 546; b) A. Fernµndez-Mateos,
S. E. Madrazo, P. H. Teijón, R. R. Gonzµlez, Eur. J. Org. Chem.
2010, 856.
[19] CCDC 1402061 (14), CCDC 1404087 (3b), CCDC 1404088 (20),
CCDC 1404089 ((À)-1) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre. The
solvent(s) used for crystallization to generate the crystals (20)
were dichloromethane and petroleum. Given that the compo-
nents in petroleum are complex and that there are large cavities
in the crystals, residual solvent molecules were obviously
disordered and the results reported mainly depended on the
positions of electron density peaks on the difference Fourier map
and the structure refinement. Since compound 14 was optically
pure (see page 60 in the Supporting Information), chiral
inversion or racemization could not take place in the following
procedure, and it is believed that the Flack constant which is not
equal to 0 for (À)-1, was caused by experimental error.
[23] S. Dua, M. S. Taylor, M. A. Buntine, J. H. Bowie, J. Chem. Soc.
Perkin Trans. 2 1997, 1991.
[24] It should be noted that the highly reactive proton on C9 of 16 is
both a to the carbonyl and benzylic, and is also exposed on the
exo-face of the concave skeleton, thus making it easy to
deprotonate by methylating reagents to form an enolate, which
would reform the starting material upon quenching the reaction.
[20] C. S. Carman, G. F. Koser, J. Org. Chem. 1983, 48, 2534.
´
[21] M. Mortimore, G. S. Cockerill, P. Kocienski, R. Treadgold,
Received: July 16, 2015
Tetrahedron Lett. 1987, 28, 3747.
Published online: September 14, 2015
Angew. Chem. Int. Ed. 2015, 54, 13599 –13603
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 13603