Synthesis of glaucogenin D, a structurally unique disecopregnane steroid with potential antiviral activity

Article abstract of DOI:10.1021/ol402193b

The first chemical synthesis of glaucogenin D, a 13,14:14,15-disecopregnane steroid with potential antiviral activity, has been accomplished in 12 steps from a hirundigenin-type intermediate. The present route would also be amenable to the synthesis of na

Full text of DOI:10.1021/ol402193b

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of Glaucogenin D, a Structurally
Unique Disecopregnane Steroid with
Potential Antiviral Activity
Jinghan Gui,* Hailong Tian, and Weisheng Tian*
Key Laboratory of Synthetic Chemistry of Natural Substances, Shanghai Institute of
Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road, Shanghai, China
guijh@sioc.ac.cn; wstian@sioc.ac.cn
Received August 2, 2013
ABSTRACT
The first chemical synthesis of glaucogenin D, a 13,14:14,15-disecopregnane steroid with potential antiviral activity, has been accomplished in
12 steps from a hirundigenin-type intermediate. The present route would also be amenable to the synthesis of natural and unnatural glaucogenin
derivatives for SAR studies.
Natural products and their derivatives have long been
important sources for drug discovery and development.1
A countless variety of terrestrial and marine organisms
produce a rich diversity of natural products, constantly
presenting new structural features that elicit chemical
and biological curiosity. Glaucogenins (Figure 1, 1À5)
are steroids with unusual structures of 13,14:14,15-diseco-
pregnane from the asclepiadaceae plant, typical of three
Chinese herbal medicines “Bai-Qian”, “Bai-Wei”, and
“Xu-Chang-Qing” from the 1980s.2 Structurally, the glau-
cogenins possess a characteristic nine-membered lactone C
ring, bis-furan cis-fused D/E rings, and polyhydroxylated
A/B rings. The unique features of the C, D, and E rings
make the glaucogenins appealing targets for synthetic
Figure 1. Structures of glaucogenin family.
chemists. Importantly, glaucogenin C (3) was found to
display effectiveand selective inhibitory activity toR-virus-
like positive-strand RNA viruses (IC50 = 17 nm),3
although the activities of other members of this family
remain to be elucidated. The intriguing chemical structures
as well as the limited biological data of the glaucogenin
family prompted us to initiate a synthetic project aiming to
offer efficient access to all glaucogenin members and their
analogues, togetherwithevaluating their potential asnovel
antiviral drugs.
(1) (a) Harvey, A. L. Drug Discov. Today 2008, 13, 894. (b) Li,
J. W.-H.; Vederas, J. C. Science 2009, 325, 161. (c) Newman, D. J.;
Cragg, G. M. J. Nat. Prod. 2012, 75, 311.
(2) (a) Nakagawa, T.; Hayashi, K.; Mitsuhashi, H. Tetrahedron Lett.
1982, 23, 757. (b) Nakagawa, T.; Hayashi, K.; Mitsuhashi, H. Chem.
Pharm. Bull. 1983, 31, 870. (c) Nakagawa, T.; Hayashi, K.; Mitsuhashi,
H. Chem. Pharm. Bull. 1983, 31, 2244. (d) Nakagawa, T.; Hayashi, K.;
Mitsuhashi, H. Chem. Pharm. Bull. 1983, 31, 879. (e) Nakagawa, T.;
Hayashi, K.; Wada, K.; Mitsuhashi, H. Tetrahedron 1983, 39, 606. (f)
Zhang, Z.; Zhou, J.; Hayashi, K.; Mitsuhashi, H. Chem. Pharm. Bull.
1985, 33, 1507. (g) Zhang, Z. X.; Zhou, J.; Hayashi, K.; Mitsuhashi, H.
Chem. Pharm. Bull. 1985, 33, 4188. (h) Sugama, K.; Hayashi, K.;
Mitsuhashi, H.; Kaneko, K. Chem. Pharm. Bull. 1986, 34, 4500. (i)
Dou, J.; Bi, Z.-M.; Zhang, Y.-Q.; Li, P. Chin. J. Nat. Med. 2006, 4, 192.
(j) Zhang, M.; Wang, J.-S.; Luo, J.; Wei, D.-D.; Kong, L.-Y. Nat. Prod.
Res. 2013, 27, 176.
Previously, we have studied the biomimetic synthesis of
a key disecopregnane skeleton from hirundigenin-type
intermediate 6, which allowed us to complete the first
(3) Li, Y.; Wang, L.; Li, S.; Chen, X.; Shen, Y.; Zhang, Z.; He, H.;
Xu, W.; Shu, Y.; Liang, G.; Fang, R.; Hao, X. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 8083.
r
10.1021/ol402193b
XXXX American Chemical Society

With a powerful approach to construct the nine-
membered lactone ring in hand, efforts were then directed
toward the synthesis of natural glaucogenins. Through
inspection of the structures of the glaucogenin family
(Figure 1), one could easily observe that they differ only
in the oxidation patterns at C1, C2 and C7. As such, we
postulated that a unified synthetic strategy6 with enone 14
as the key intermediate would enable access to the entire
glaucogenin family (Scheme 1B): reduction or deoxygena-
tion of enone 14 would provide glaucogenin D or C, while
further introduction of hydroxyl groups at C1 or C2via the
C3-carbonyl group should afford glaucogenin A, B, or E.
Herein, we report the first synthesis of glaucogenin D (4),
which represents an entry toward the unified synthesis of
the glaucogenin family.
Scheme 1. (A) Construction of the Disecopregnane Skeleton
from Hirundigenin-Type Intermediates via Regioselective
Radical Fragmentation Reactions. (B) Unified Synthetic
Strategy toward the Glaucogenin Family
As shown in Scheme 2, our synthesis of glaucogenin D
(4) commenced with a diastereoselective cis-dihydroxyla-
tion of 12 with KMnO47 to furnish diol 15 in 70% yield.
Oxidation of the C7-hydroxyl group led to ketone 16,
which was primed for the dehydrogenation event to install
the critical C5ÀC6 olefin. Initial attempts to effect this
dehydrogenation directly with DDQ8 or IBX9 met with
failure. Alternatively, ketone 16 reacted with LDAÀ
TMSCl to afford the corresponding silyl enol ether 17,
which underwent bromination to produce bromide 18 in
excellent yield. Unfortunately, dehydrobromination of 18
proved to be a dead end under a variety of conditions. We
surmised that this failure might result from the cis-config-
uration of the C6ÀBr and C5ÀH, which does not satisfy
the antiperiplanar conformation requirement of an E2
elimination. Therefore, we explored an alternative selen-
oxide syn-elimination pathway10 by converting 17 to phe-
nylseleno ketone 20 with PhSeCl (99% yield from 16).
Surprisingly and unexpectedly, when PhSeBr was used as
the selenium reagent, bromide 18 was obtained as the sole
product instead of phenylseleno ketone 20. Contrary to
literature precedent, this provides a rare example of a
reaction between a silyl enol ether and PhSeBr to generate
a bromide product.
synthesis of 5,6-dihydroglaucogenin C.4 Key features of
the previous synthesis were a Schenck ene reaction and
subsequent iron(II)-promoted regioselective fragmenta-
tion of an R-alkoxy hydroperoxide (Scheme 1A, see 6 f
8, 8 f 10, and 12). To synthesize more glaucogenin mem-
bers as well as their analogues by utilizing our methodol-
ogy, we continued to explore the reactivity of 7, an anal-
ogue of 6bearing an additional C15-acetal group. Pleasingly,
Schenck ene reaction5 of 7 also afforded the alkoxy hydro-
peroxide 9 as a single diastereomer in quantitative yield.
Hydroperoxide 9 was found to display a similar re-
activity under the optimal conditions4 we had developed
for the fragmentation reaction of 8. Upon exposure to
iron(II), 9 decomposed efficiently to give a C13-alkyl radical
that can be trapped by either TEMPO or I2 to produce,
respectively, C13-TEMPO-substituted product 13 in 71%
yield or C13ÀC18 olefin product 11 in 83% yield after
iodide elimination. This successful synthesis of disecopreg-
nane intermediate 13 would provide access to more glauco-
genin analogues with a variety of possible substituents at
C15, which could be easily realized via acetal chemistry.
Deprotection of 20 and subsequent oxidation with
ozone11 gave the corresponding selenoxide, which under-
went syn-elimination to afford enone 22 in 52% yield from
20. Fortunately, we were able to obtain a single crystal
structure of the deprotected phenylseleno ketone 21,12
which confirmed the stereochemical assignments at C6
(6) (a) Lavilla, R.; Gullon, F.; Bosch, J. Eur. J. Org. Chem. 1999, 373.
(b) Tisdale, E. J.; Slobodov, I.; Theodorakis, E. A. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 12030. (c) Dransfield, P. J.; Dilley, A. S.; Wang, S.;
Romo, D. Tetrahedron 2006, 62, 5223. (d) Broehmer, M. C.; Bourcet, E.;
Nieger, M.; Braese, S. Chem.;Eur. J. 2011, 17, 13706.
(7) Hegedus, L. S.; Geisler, L.; Riches, A. G.; Salman, S. S.;
Umbricht, G. J. Org. Chem. 2002, 67, 7649.
(8) Walker, D.; Hiebert, J. D. Chem. Rev. 1967, 67, 153.
(9) Satam, V.; Harad, A.; Rajule, R.; Pati, H. Tetrahedron 2010, 66,
7659.
(10) Nishibayashi, Y.; Uemura, S. Top. Curr. Chem. 2000, 208, 201.
(11) Bonjoch, J.; Sole, D.; Garcia-Rubio, S.; Bosch, J. J. Am. Chem.
Soc. 1997, 119, 7230.
(12) CCDC-920710 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(4) Gui, J.; Wang, D.; Tian, W. Angew. Chem., Int. Ed. 2011, 50, 7093.
(5) (a) Prein, M.; Adam, W. Angew. Chem., Int. Ed. 1996, 35, 477. (b)
Clennan, E. L. Tetrahedron 2000, 56, 9151.
B
Org. Lett., Vol. XX, No. XX, XXXX

Scheme 2. Synthesis of Glaucogenin D (4)
and C8, as well as C13, C16 and C21 from our early
13
synthetic route. Dehydroxylation of 22 at C8 with SmI2
and Luche reduction of the resulting conjugated enol
furnished C7β-OH product 23 as the major product in
58% yield.
Scheme 3. Complementary Olefin Formation from Two
Distinct C13-Substituted Disecopregane Intermediates
What remained in the synthesis of glaucogenin D (4) was
the removal of the TEMPO group in 23 to reveal the
C13ÀC18 olefin, which unluckily proved to be a nontrivial
task. Employing 12 as the model substrate (Scheme 3), we
were frustrated to find that TEMPO remained intact when
using conventional methods14 for the heterolytic cleavage
of the NÀO bond, presumably due to the severe steric
hindrance around the NÀO bond in 12 (four methyl
groups around N and a tetrasubstituted C13 atom at-
tached to O). We then drew inspiration from the thermal
decomposition of alkoxyamines inthefield of living radical
polymerization where the TEMPO group could be re-
moved to afford an olefin directly through homolytic
cleavage of the CÀO bond rather than the NÀO bond.15
Surprisingly, under thermal conditions, 12 decomposed in
a highly regioselective manner to give C12ÀC13 olefin 25
in 95% yield, which is complementary to the regioselec-
tivity observed in the thermal elimination of iodide 26
affording C13ÀC18 olefin 10. Luckily, olefin 25 could be
isomerized to enol ether 10 using RuCl2(PPh3)3 in toluene
(71% yield),16 or using the more robust Wilkinson’s
n
catalyst in PrOH (>99% yield).17 The unexpected for-
mation of the C12ÀC13 olefin in the TEMPO thermal
decomposition deserves further comment: (1) it provides a
facile approach to the synthesis of glaucogenin analogues
(13) Nicolaou, K. C.; Ellery, S. P.; Chen, J. S. Angew. Chem., Int. Ed.
2009, 48, 7140.
(14) Conditions screened: ZnÀHOAc, Mo(CO)6, 10% Pd/C,
SmI2, etc.
(16) Tang, F.; Moeller, K. D. Tetrahedron 2009, 65, 10863.
(17) Fang, F. G.; Xie, S.; Lowery, M. W. J. Org. Chem. 1994, 59,
6142.
(15) (a) Ananchenko, G. S.; Fischer, H. J. Polym. Sci. Polym. Chem
2001, 39, 3604. (b) Goto, A.; Kwak, Y.; Yoshikawa, C.; Tsujii, Y.;
Sugiura, Y.; Fukuda, T. Macromolecules 2002, 35, 3520.
Org. Lett., Vol. XX, No. XX, XXXX
C

bearing a C12ÀC13 olefin, and (2) for future total synthe-
sis endeavors, a strategic disconnection could be reason-
ably made via cleavage of the central nine-membered
lactone at the C14 ester bond or the C12ÀC13 olefin.
Finally, a two-step sequence as shown in Scheme 2,
including removal of the TEMPO group in 23 and iso-
merization, delivered synthetic glaucogenin D (4), whose
1H and 13C NMR spectra were in good agreement with
that of natural glaucogenin D (4).2i
structural diversification toward glaucogenin analogues,
including but not limited to C15-acetal, C12ÀC13 olefin,
and various oxidation states of the B ring, which are of
significant importance to the pharmaceutical research of
disecopregnane steroids. Additionally, we anticipate that
allylic deoxygenation of glaucogenin D (4) would provide
glaucogenin C (3), and further oxidation of the A-ring
might realize the synthesis of other glaucogenin members,
which is currently being pursued in our laboratory.
In conclusion, we have completed the first chemical
synthesis of glaucogenin D (4) in 12 steps and 6.8% overall
yield starting from hirundigenin-type intermediate 6.
Salient features of the current synthesis include (1) estab-
lishment of the requisite C5ÀC6 olefin via a thermal syn-
elimination reaction of a phenyl selenoxide; (2) discovery
of thermal conditions to remove the TEMPO group to
give a complementary C12ÀC13 olefin that can be either
isomerized to a C13ÀC18 olefin or deconstructed for
future total synthesis endeavors; (3) the possibility of
Acknowledgment. We gratefully acknowledge financial
support from the National Science Foundation of China
(21290180).
Supporting Information Available. Experimental pro-
1
cedures, spectroscopic data, and copies of H and 13C
NMR spectra. This material is available free of charge via
the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX