Design, synthesis, and biological evaluation of truncated superstolide A

Article abstract of DOI:10.1002/anie.201209300

By design: A truncated superstolide A, a simplified analogue of the potent anticancer marine macrolide superstolide A, was designed and successfully synthesized by a highly efficient and convergent approach. The biological evaluation showed that this compound maintains the potent anticancer activity of the original natural product superstolide A.

Full text of DOI:10.1002/anie.201209300

.
Angewandte
Communications
DOI: 10.1002/anie.201209300
Medicinal Chemistry
Design, Synthesis, and Biological Evaluation of Truncated
Superstolide A**
Lei Chen, Kausar Begam Riaz Ahmed, Peng Huang, and Zhendong Jin*
Dedicated to Professor Philip L. Fuchs on the occasion of his 68th birthday
Marine macrolides are well known for their fascinating
molecular structure and potent anticancer activity.[1] Super-
stolides A (1) and B (2; Scheme 1), two marine macrolides,
were isolated in minute amounts from the deep-water marine
sponge Neosiphonia superstes.[2] Their absolute structures
potential to cause significant damage to the marine habitat.
Furthermore, sponges are notorious for being extremely
difficult animals to cultivate in controlled systems.[3] The lack
of an adequate compound supply severely impedes research
toward a thorough understanding of their mechanism of
action and hampers the gain of further insight into the clinical
potential of this group of fascinating marine macrolides.
These compounds have attracted a great deal of attention
from the synthetic organic chemistry community because of
their potent anticancer activities coupled with their challeng-
ing molecular structures.[4,5] Roush and his co-workers have
reported the only completed total synthesis of superstol-
ide A.[5h]
Several years ago, we initiated research directed toward
the total synthesis of superstolides A and B.[4] While working
on the total synthesis, it became apparent to us that because of
the structural complexity of the target molecules, it would be
extremely challenging to develop a practical total synthesis
that is capable of providing an adequate amount of material
for biological investigation, therapeutic evaluation, and
possible future clinical trials.
Owing to the lack of a sufficient amount of these natural
products and the overwhelming difficulty in the development
of a practical total synthesis approach we decided to design
a simplified superstolide A analogue that contains the basic
pharmacophore and can be easily synthesized in a much
shorter reaction sequence. Herein, we report for the first time
the design and synthesis of a truncated superstolide A (3), in
which the cis-fused functionalized decalin is simplified to
a cyclohexene ring, whereas the 16-membered macrolactone
remains intact (Scheme 2).
This design is based on our hypothesis that the 16-
membered macrolactone may be the key pharmacophore that
interacts with cellular target(s), whereas the cis-fused decalin
may lock the macrolide into an advantageous conformation.
This modification would simplify the synthesis substantially,
and at the same time maintain the basic template of the
molecule. Such a strategy was considered important in that it
could test our hypothesis on the interaction between the
Scheme 1. Anticancer marine macrolide superstolides A and B.
were determined by using extensive spectroscopic methods.
The structural novelty of these two molecules is characterized
by a unique 16-membered macrolactone attached to a func-
tionalized cis-decalin.
Both superstolides A and B exhibit a potent antiprolifer-
ative effect against several tumor cell lines with IC50 values
ranging from 4.8 to 64 nm.[2] As they have novel and
unprecedented structures we can deduce that they might
have a unique cellular target(s) and a novel mechanism of
action. Unfortunately, the yields of isolation for superstol-
ides A and B are only 0.003% and 0.0003%, respectively. In
addition, the marine sponge Neosiphonia superstes lives at
500–515 meters deep in the ocean off New Caledonia, thus
making the collection of a large amount of the marine sponge
very difficult and dangerous and its collection has the
[*] Dr. L. Chen, Prof. Dr. Z. Jin
Division of Medicinal and Natural Products Chemistry
Department of Pharmaceutical Sciences & Experimental Thera-
peutics
College of Pharmacy, The University of Iowa
Iowa City, IA 52242 (USA)
E-mail: zhendong-jin@uiowa.edu
K. B. Riaz Ahmed, Prof. Dr. P. Huang
Department of Molecular Pathology
The University of Texas MD Anderson Cancer Center
Houston, TX 77030 (USA)
[**] This work was made possible by the generous support of the NIH
(5R01CA109208).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209300.
Scheme 2. Design of truncated superstolide A.
3446
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3446 –3449

Angewandte
Chemie
natural product and the receptor and provide important
information regarding the structure–activity relationship and
pharmacophore identification. In addition, the synthesis of
truncated superstolide A (3) would also serve as an important
model study that would provide some critically important
information on the feasibility of three key coupling reactions
and an ester formation (or macrolactonization) in our total
synthesis strategy especially considering the enormous diffi-
culty involved in the macrolactonization step encountered in
the synthesis of Rousch and co-workers.[5i]
Scheme 3 outlines the retrosynthetic analysis of truncated
superstolide A (3). Sequential disconnections reveal frag-
ments 4, 5, and 6 as potential key intermediates, with Suzuki,
Negishi, and Stille couplings playing crucial roles in the
synthetic strategy.
Scheme 4. Enantioselective synthesis of fragment 4. binol=2,2’-dihy-
droxy-1,1’-binaphthyl, DIBAL=diisobutylaluminium hydride, DMSO=
dimethyl sulfoxide, HMPA=hexamethylphosphorylamide, LDA=
lithium diisopropylamide.
Scheme 3. Retrosynthetic analysis of truncated superstolide A (3).
TBS=tert-butyldimethylsilyl, TES=triethylsilyl.
Our starting material lactone 8 was prepared enantiose-
lectively in 73% yield (95% ee) by using an elegant Diels–
Alder reaction developed by Ward and Santos (Scheme 4).[6]
Lactone 8 was treated with LDA and the the resulting enolate
was quenched with MeI to provide compound 11 in 95% yield
and with the requisite stereochemistry at the quaternary
carbon. Reduction with DIBAL to lactol 12 and subsequent
addition of lithiotrimethylsilyldiazomethane gave alkyne 7 in
88% yield.[7] Compound 7 was treated with 2.5 equivalents of
nBuLi to afford a dianion, which was quenched with
3 equivalents of TESOTf to furnish an intermediate. This
intermediate was treated with 5% HCl to chemoselectively
cleave the TES ether to provide compound 13 in 86% yield.
Homoallylic alcohol 13 was carefully oxidized to its corre-
sponding aldehyde 14, which was immediately converted into
geminal dibromo compound 4 in 72% yield.
Scheme 5. Synthesis of fragment 5. Red-Al=sodium bis(2-methoxy-
ethoxy)aluminum hydride.
Hydrolysis of compound 18 furnished carboxylic acid 5 in
78% yield.
The cross-metathesis between olefin 19[4a] and pinacol
vinylboronate 20 proved to be quite challenging (Scheme 6).
After much investigation, it was discovered that when the
second generation of Grubbs-Hoveyda catalyst was used
compound 19 was successfully converted into trans-vinyl-
boronate 6 in 83% yield.[9]
Compound 5 was synthesized using a modified literature
The Suzuki coupling between geminal dibromo com-
pound 4 and vinyl boronate 6 provided compound 21 in 70%
yield with complete stereoselectivity (Scheme 7).[10] Negishi
coupling between vinyl bromide 21 and Me2Zn gave the
procedure (Scheme 5).[5f] Compound 16[8] was oxidized to
aldehyde 17, which underwent
a Horner–Wadsworth–
Emmons olefination to give compound 18 in 89% yield.
Angew. Chem. Int. Ed. 2013, 52, 3446 –3449
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3447

.
Angewandte
Communications
Scheme 6. Synthesis of fragment 6.
requisite trisubstituted olefin 22 in 86% yield with complete
stereoselectivity.[11] It should be noted that the triethylsilyl
group attached to the alkyne moiety of compound 21 was
extremely important because it prevented the facile cyclic
carbopalladation and subsequent cross-coupling, a major side
reaction that competed with the desired Negishi coupling.[12]
Three silyl protecting groups were removed by TBAF to
afford alkyne 23, which underwent a regio and stereoselective
hydrostannylation to furnish vinyl stannane 24 in 69% yield.
A regioselective esterification between alcohol 24 and
carboxylic acid 5 provided compound 25, which isomerized
to the desired ester 26 upon treatment with Ti(OiPr)4. This
remarkable acyl migration was first reported by Paterson and
Mackay.[5f] The formation of the thermodynamically more-
stable compound 26 is presumably due to the additional
hydrogen bond between the amide group and the neighboring
hydroxy group in 26. Finally, compound 26 underwent a facile
intramolecular Stille coupling to give truncated superstol-
ide A (3) in 88% yield.[13]
In summary, the truncated superstolide A (3) was success-
fully synthesized in 15 steps from commercially available
starting material 9 in 6.2% overall yield.[14] It should be noted
that all stereogenic carbon centers and double bonds present
in the target molecule were constructed stereoselectively.
The antiproliferative effect of compound 3 was evaluated
in eight cancer cell lines by using the MTT (3-(4,5-di-
Scheme 7. Coupling of the three major fragments and completion of
the synthesis. dba=dibenzylideneacetone, DMAP=4-dimethylamino-
pyridine, DMF=N,N’-dimethylformamide, EDCI=1-ethyl-3-(3’-di-
methylaminopropyl)carbodiimide hydrochloride, TBAF=tetra-n-buty-
lammonium fluoride.
methylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide)
assay.[15] The IC50 data are shown in Table 1. We were
delighted to discover that compound 3 is about seven times
more potent in suppressing tumor cell proliferation than its
parent natural product superstolide A in HT-29 cell (the IC50
value for superstolide A in HT-29 is 64 nm).[2a] In addition,
compound 3 is also potent in suppressing tumor cell
proliferation in the other seven tested cell lines with IC50
values ranging from 12–77 nm. These results have confirmed
our hypothesis that the 16-membered macrolactone is indeed
the pharmacophore that interacts with its putative target in
the cells, and the modification of the fuctionalized cis-decalin
to a cyclohexene ring apparently does not affect its potent
anticancer activity. This information is highly valuable for the
design of other analogues in the future.
In conclusion, we have designed and synthesized a trun-
cated superstolide A (3) that maintains the potent anticancer
activity of the original natural product superstolide A. There-
fore, we have successfully indirectly solved the supply
problem of superstolide A. In addition, our highly convergent
and flexible synthetic approach can be easily modified for the
synthesis of other designed analogues and bioconjugates in
Table 1: Antiproliferative effect of compound 3 on various malignant
tumor cells (MTT assay).
Entry
Cell Line
IC50 [nm]
1
2
3
4
5
6
7
8
HT29
7.54
36.52
53.06
63.82
56.75
52.71
76.73
11.85
A375SM
MEL624
SK-MEL-2
H1299
HCT116
Raji
HL60
3448
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3446 –3449

Angewandte
Chemie
J. J. Mulhearn, Org. Lett. 2005, 7, 1593; h) M. Tortosa, N. A.
the future. Research in this direction is currently underway
and will be reported in due course.
Yakelis, W. R. Roush, J. Am. Chem. Soc. 2008, 130, 2722; i) M.
Tortosa, N. A. Yakelis, W. R. Roush, J. Org. Chem. 2008, 73,
9657.
Received: November 20, 2012
Revised: December 27, 2012
Published online: February 13, 2013
[6] D. E. Ward, M. S. Santos, Org. Lett. 2005, 7, 3533.
[7] S. Ohira, K. Okai, T. Moritani, J. Chem. Soc. Chem. Commun.
1992, 721.
[8] Y. Hayashi, M. Shoji, H. Ishikawa, J. Yamaguchi, T. Tamura, H.
Imai, Y. Nishigaya, K. Takabe, H. Kakeya, H. Osada, Angew.
Chem. 2008, 120, 6759; Angew. Chem. Int. Ed. 2008, 47, 6657.
[9] a) C. Morrill, R. H. Grubbs, J. Org. Chem. 2003, 68, 6031; b) C.
Morrill, T. W. Funk, R. H. Grubbs, Tetrahedron Lett. 2004, 45,
7733.
Keywords: anticancer agent · asymmetric synthesis ·
drug design · macrolide · superstolide analogue
.
[1] a) Y. Qi, S. Ma, ChemMedChem 2011, 6, 399, and references
therein; b) I. Bhatnagar, S.-K. Kim, Mar. Drugs 2010, 8, 2702,
and references therein.
[10] A. Suzuki, J. Organomet. Chem. 1999, 576, 147, and references
therein.
[2] a) M. V. DꢀAuria, C. Debitus, L. G. Paloma, L. Minale, A.
Zampella, J. Am. Chem. Soc. 1994, 116, 6658; b) M. V. DꢀAuria,
C. Debitus, L. G. Paloma, L. Minale, A. Zampella, J. Nat. Prod.
1994, 57, 1595.
[11] E.-i. Negishi, Q. Hu, Z. Huang, M. Qian, G. Wang, Aldrichimica
Acta 2005, 38, 71, and references therein.
[12] E.-i. Negishi, Y. Noda, F. Lamaty, E. J. Vawter, Tetrahedron Lett.
1990, 31, 4393.
[3] S. Sagar, M. Kaur, K. P. Minneman, Mar. Drugs 2010, 8, 2619,
and references therein.
[4] a) W. Yu, Y. Zhang, Z. Jin, Org. Lett. 2001, 3, 1447; b) Z. Hua, W.
Yu, M. Su, Z. Jin, Org. Lett. 2005, 7, 1939; c) L. Chen, Z. Hua, G.
Li, Z. Jin, Org. Lett. 2011, 13, 3580.
[5] a) W. R. Roush, J. A. Champoux, B. C. Peterson, Tetrahedron
Lett. 1996, 37, 8989; b) A. Zampella, M. V. DꢀAuria, Tetrahe-
dron: Asymmetry 2001, 12, 1543; c) W. R. Roush, L. Hertel, M. J.
Schnaderbeck, N. A. Yakelis, Tetrahedron Lett. 2002, 43, 4885;
d) N. A. Yakelis, W. R. Roush, J. Org. Chem. 2003, 68, 3838;
e) J. G. Solsona, P. Romea, F. Urpi, Org. Lett. 2003, 5, 4681; f) I.
Paterson, A. C. Mackay, Synlett 2004, 1359; g) J. A. Marshall,
[13] a) J. K. Stille, Angew. Chem. 1986, 98, 504; Angew. Chem. Int.
Ed. Engl. 1986, 25, 508, and references therein; b) G. Pattenden,
D. J. Sinclair, J. Organomet. Chem. 2002, 653, 261, and references
therein; c) K. C. Nicolaou, P. G. Bulger, D. Sarlah, Angew.
Chem. 2005, 117, 4516; Angew. Chem. Int. Ed. 2005, 44, 4442, and
references therein.
[14] As 11 steps were needed to prepare compound 19 from (S)-(À)-
ethyl lactate, the longest linear sequence for the synthesis of
compound 3 consists of 19 steps with 3.2% overall yield.
[15] The procedure for cell viability assay and MTT curves are
included in the Supporting Information.
Angew. Chem. Int. Ed. 2013, 52, 3446 –3449
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3449