Total synthesis and biological evaluation of (-)-englerin A and B: Synthesis of analogues with improved activity profile

Article abstract of DOI:10.1002/anie.201007790

The large-scale synthesis of englerin A (see scheme) and subsequent structure-activity relationship studies have led to the discovery of highly potent analogues. TBS=tert-butyldimethylsilyl.

Full text of DOI:10.1002/anie.201007790

Communications
DOI: 10.1002/anie.201007790
Natural Products
Total Synthesis and Biological Evaluation of (À)-Englerin A and B:
Synthesis of Analogues with Improved Activity Profile**
Lea Radtke, Matthieu Willot, Hongyan Sun, Slava Ziegler, Stephanie Sauerland,
Carsten Strohmann, Roland Frꢀhlich, Peter Habenberger, Herbert Waldmann, and
Mathias Christmann*
Dedicated to Professor Dieter Enders on the occasion of his 65th birthday
Although successful in many types of cancer, chemotherapy
has shown at best moderate effectiveness in the treatment of
renal cell carcinoma.[1] Currently approved therapies target
the downstream signaling that leads to an over-expression of
angiogenic factors such as VEGF and PDGF and include
tyrosine kinase inhibitors (sorafenib, sutinib) as well as
mTOR inhibitors (temsirolimus, everolimus). Owing to
severe side effects and other drawbacks of the above-
mentioned treatments, the identification of novel inhibitors
Scheme 1. Strategy for the diverted synthesis of englerin A derivatives
of kidney cancer signaling pathways remains highly desirable.
Toward this end, a collection of plant extracts was screened
against an NCI 60-cell panel containing renal cancer cell lines
along with eight other organ panels. Selecting for specific
inhibitors of renal cancer cell lines led to the identification of
englerin A and B, metabolites of Phyllanthus engleri, which is
a plant indigenous to the East African countries of Tanzania
and Zimbabwe. Englerin A (1) is a densely functionalized
guaiane sesquiterpenoid with an oxatricyclic core flanked by
two opposing ester side chains (Scheme 1). Following Beu-
tlerꢀs initial report[2] on the strong and selective activity of 1
against six of eight renal cancer cell lines, both compounds
from intermediate 2 (green box) or earlier intermediates (red box).
TBS=tert-butyldimethylsilyl.
received immediate attention from the synthetic community.
In 2009, our research group established the previously
unknown absolute configuration of englerin A by total syn-
thesis of its (+)-enantiomer.[3] Shortly after, the groups of
Ma[4] and Echavarren[5] reported independent total syntheses
of (À)-1 using elegant gold-catalyzed cyclization cascade
reactions of open-chain precursors. A fourth synthesis of
englerin was published by Nicolaou, Chen, and co-workers[6]
and features an intermolecular rhodium-catalyzed
[4+3] cycloaddition strategy and progress reports from
other groups[7] are harbingers of ongoing interest.
Herein, we report a reliable and scalable route to
englerinꢀs natural enantiomer (À)-1 via intermediate 2,
encompassing englerinꢀs complete guaiane core (2). This
material fuelled an extensive SAR study[8] aimed to clarify the
role of englerinꢀs two ester side chains (green boxes).
Diverting the synthesis[9] at an earlier intermediate allowed
us to probe the influence of the isopropyl group (red box).
When devising a strategy for a total synthesis program, it is
mandatory to plan for the preparation of both enantiomers
when the absolute configuration is unknown. The monoter-
pene nepetalactone possesses an iridoid substructure that is
primed for olefin oxidation leading to englerinꢀs oxidation
pattern. While (+)-nepetalactone is readily available in bulk
quantities from essential oil companies, the (À)-enantiomer 3
can be synthesized from (+)-citronellal on multigram scale
following Schreiberꢀs protocol.[10] In our initial total synthesis,
the absolute configuration of englerin A was deduced from
natural (+)-nepetalactone. For the biological evaluation of
the natural product and key analogues, we had to start from
synthetic nepetalactone 3 and thus taking advantage of the
previously developed chemistry.[3a] This material was con-
[*] L. Radtke, Dr. M. Willot, S. Sauerland, Prof. C. Strohmann,[+]
Prof. H. Waldmann, Prof. M. Christmann
TU Dortmund University, Faculty of Chemistry
Otto-Hahn-Strasse 6, 44227 Dortmund (Germany)
Fax: (+49)231-755-5363
E-mail: mathias.christmann@tu-dortmund.de
Homepage: http://www.chemie.tu-dortmund.de/christmann
Dr. R. Frꢀhlich[+]
University of Mꢁnster, Institute of Organic Chemistry
Corrensstrasse 40, 48149 Mꢁnster (Germany)
Dr. H. Sun, Dr. S. Ziegler, Prof. H. Waldmann
Department of Chemical Biology
Max-Planck-Institute of Molecular Physiology
Otto-Hahn-Strasse 11, 44227 Dortmund (Germany)
Dr. P. Habenberger
Lead Discovery Center GmbH
Emil-Figge-Strasse 76a, 44227 Dortmund (Germany)
[+] X-ray crystal structure analysis.
[**] We thank the Fonds der Chemischen Industrie (Dozentenstipen-
dium to M.C.), the Alexander von Humboldt Foundation for
postdoctoral fellowships (to M.W. and H.S.) and finally, we thank
Takasago International Corporation for generous donation of (+)-
citronellal.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201007790.
3998
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3998 –4002

The major challenge in the remainder of the synthesis was
the improvement of the moderate diastereoselectivity for the
olefin epoxidation reported previously. Although the aceto-
nide moiety exerted a very favorable influence on the desired
facial selectivity (d.r. 11:1, see below), subsequent cleavage of
the acetonide group in the presence of the epoxide group led
to extensive decomposition. We therefore removed the
acetonide unit prior to the epoxidation and studied the
directing effect of several residues on the secondary alcohol
(Scheme 3).
Scheme 2. Diverting the synthesis from bicyclic aldehyde 4 using zinc-
mediated allylation reactions. THF=tetrahydrofuran.
verted in decagram scale to the early key intermediate 4 and
subjected to Barbier-type allylations (Scheme 2). At this
juncture, we also opted for three different allyl bromides to
investigate the role of the isopropyl substituent for the
biological activity of the natural product. Reduction of the
lactone (5a–c) with LiAlH4 afforded the triols 6a–c in
excellent yields. The main portion of 4 was converted into
the isopropyl derivative 5a. Interestingly, the minor diaste-
reomer of 5a can be recycled in part by oxidation with IBX to
the ketone and subsequent diastereoselective reduction of
both the ketone and the lactone moiety (LiAlH4, d.r. 3:1) to
give crystalline triol intermediate 6a.
In turn, the triols 6a–c were converted into the dienes 7a–
c, which are precursors for the crucial ring-closing metathesis
reaction. In the original total synthesis, we used 20 mol% of
the second-generation Grubbs catalyst (Grubbs II) to achieve
complete conversion of 7a (R = iPr) into the seven-mem-
bered ring with trisubstituted double bond. This amount could
be lowered to 15 mol% without compromising the yield,
whereas lower catalyst loadings led to incomplete conver-
sions. Interestingly, when the size of the R group was
decreased (iPr> Et > Me), the rate of the ring-closing meta-
thesis increased dramatically leading to excellent turnover
and yields with as little as 5 mol% of catalyst (see Table 1).
Scheme 3. Removal of the acetonide group and protection of the
hydroxy group(s).
Early introduction of the TBS-protected glycolate side
(10a) led to a directed epoxidation in favor of desired
diastereomer 11a in good yield albeit with a moderate 2.3:1
ratio (Table 2, entry 1). Switching to the monoprotected TBS
Table 2: Influence of the protecting group pattern on the diastereo-
selectivity of the epoxidation with m-CPBA.
Entry
1
Alkene
R1
H
R2
Epoxide
11/12
Yield [%][a]
90
10a
11a,12a
2.3:1
2
3
4
5
6
10b
10c
10d
8a
H
TBS
H
1,2-acetonide
H
TBS
TBS
Troc
11b,12b
11c,12c
11d,12d
11e,12e
11 f,12 f
5.4:1
9:1
1:20
11:1
1:3
91
99
73
78
99
Table 1: Synthesis of the englerin’s guaiane core using a ring-closing
metathesis (RCM) reaction.
9a
H
[a] Yield of isolated product. Troc=trichloroethoxycarbonyl.
derivative 10b afforded a significant increase in selectivity
(11b/12b = 5.4:1; Table 2, entry 2) without compromising the
yield. This trend could be enhanced further with TBS groups
on both alcohols (9:1, 99% yield; Table 2, entry 3). Most
strikingly, protection of the secondary alcohol with the Troc
group (Table 2, entry 4) resulted in a complete inversion of
the facial selectivity. Despite the excellent selectivity of the
acetonide derivative 8a (Table 2, entry 5), problems with the
subsequent conversions did not render this a viable alter-
native. Finally, epoxidation of the free diol 9a (Table 2,
entry 6) led to the preferred formation of the undesired
diastereomer 12 f. The observed selectivities can be rational-
Entry
7: R
Catalyst (mol%)
t [h]
Yield [%][a]
1
2
3
4
5
6
7
8
9
10
7a: iPr
7a: iPr
7a: iPr
7a: iPr
7a: iPr
7a: iPr
7b: Et
7b: Et
7c: Me
7c: Me
Hoveyda–Grubbs (20)
Grubbs II (20)
Grubbs II (18)
Grubbs II (15)
Grubbs II (10)
Grubbs II (10)
Grubbs II (15)
Grubbs II (10)
Grubbs II (10)
Grubbs II (5)
24
36
60
60
8
24
60
96
18
23
0
99
99
99
46[b]
82[b]
88
82[b]
98
99
[a] Yield of isolated product. [b] Incomplete conversion.
Angew. Chem. Int. Ed. 2011, 50, 3998 –4002
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3999

Communications
ized with the interplay of shielding one of the diastereotopic
faces and the ability to coordinate the m-CPBA reagent and
direct it to the same face.
configuration.[11] Finally, the glycolate ester side chain was
introduced under standard conditions (DMAP, Et3N,
CH2Cl2), and subsequent removal of the TBS ether with
TBAF concluding our second-generation synthesis of engler-
in A (1).
For the completion of the synthesis we selected epoxide
11b because it easily undergoes a transannular epoxide
opening to form the highly crystalline key intermediate 2,
suitable for the subsequent introduction of different ester side
chains (Scheme 4). In fact, we prepared 7.8 g of this material
that served as base camp for the exploration of ester
analogues of englerin A. As showcased below for the syn-
thesis of the englerin A and B (see below), this robust and
reliable route from 2 has allowed for the synthesis and
evaluation of over 30 different ester derivatives.
The chemistry described above enables the preparation of
multigram quantities of englerin A, thus making total syn-
thesis a viable method for sustainable supply for our ongoing
biological studies. As the next step, we initiated structure-
activity relationship (SAR) studies with two goals: 1) to
investigate whether the potent activity of englerin A could be
further enhanced by modification of the ester side chains or
the isopropyl group, and 2) to identify sites within the
molecule that are less sensitive to loss of bioactivity upon
modification with the intention of rendering englerin A a tool
for chemical biology research. The cytotoxicity of 32 engler-
in A derivatives (Scheme 6) was tested with the A498 kidney
cancer cell line. This cell line was selected because of its high
sensitivity towards englerin A as described by Beutler and co-
workers.[2]
Comparison of the two synthesized enantiomers con-
firmed that (À)-1 is highly cytotoxic with an IC50 value of
45 nm while the enantiomer (+)-1 is inactive up 1 mm.
Replacing the isopropyl group with an ethyl group (13) or a
methyl group (14) resulted in a 20-fold (Et) and 100-fold (Me)
decrease of activity. We then turned our attention to the
cinnamic ester side chain. Most strikingly, we were able to
identify three derivatives with approximately twice the
potency of the parent natural product englerin A (1). The 3-
cyclohexyl acrylate 15, the 3-naphthoate 16, and the 3-
methylcinnamate derivative 17 indicate that this part of the
molecule is very responsive to structural changes in a positive
sense. Substitutions in the aromatic ring of the cinnamate
residue resulted in a decreased activity (18–25). Interestingly,
benzoates were active down to the low micromolar range (27–
31). The glycolate ester domain proved to be extremely
sensitive toward modification. Methylation of the hydroxyl
group (36) afforded a 14-fold decrease in activity. Esterifica-
tion at the same position retains some activity; however it is
unclear whether the ester 37 is rather a prodrug. Despite the
diminished activity, derivatives 38 and 39 offer an alkyne
handle for the future attachment of molecular probes. Other
derivatives (40–44) did not show activities in the nanomolar
range. The dose-response curves for the four most potent
compounds including englerin A are shown in Figure 1.
The three most active derivatives of englerin (15–17) show
activity within the same magnitude. To select for the most
promising candidate for future derivatizations, we ran viabil-
ity assays with four other cell lines (Table 3). While none of
Scheme 4. Intramolecular epoxide opening to the key intermediate 2
for the synthesis of different diesters. DMAP=4-dimethylaminopyri-
dine, TBAF=tetra-n-butylammonium fluoride.
Esterification of 2 with cinnamic acid (or other acids for
the synthesis of analogues) using a Yamaguchi esterification
was followed by cleavage of the TBS ether mediated by
TBAF to give englerin B as a crystalline solid (Scheme 5). In
the enantiomeric series, we obtained single crystals that
allowed for the unambiguous determination of the absolute
Table 3: Assessing the viability (mm) in four cell lines.
Entry
Compound
Cell types
HeLa
Caki-1
HEK293
MDA-MB-468
[a]
1
2
3
4
1
–
24.8
29.2
>30
13.9
15.9
>30
>30
15.2
17.7
>30
>30
19.4
[a]
15
16
17
–
[a]
–
[a]
–
Scheme 5. Completion of the synthesis of englerin A and B as blue-
print for the synthesis of analogues.
[a] Inactive, no IC50 value was determined.
4000 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3998 –4002

Figure 1. Cytotoxicity dose-response curves are shown for 15, 16, and
17 in comparison to englerin A (À)-1. A498 cells were treated in four
replicates with different concentration of the compounds for 48 h.
Cytotoxic activity was determined using the WST-1 reagent. Data are
shown as mean (n=4) ÆSD and were fitted using GraFit 5.0
software.
cell lines, rendering it a highly potent and selective inhibitor
of the A498 kidney cancer cell line.
To investigate whether englerin and its active analogues
are able to selectively target kidney cancer cell lines we
investigated the cytotoxicity of compounds 1 and 15–17 in
several normal kidney cell lines of different origin (MDCK,
BSC-1, RC-124). To our delight the tested compounds
differentiate between them and renal cancer cell lines and
are more active on the cancer cell lines by at least a factor of
100 (with IC50 > 5 mm). Such selectivity is rare among
potential candidate compounds for drug discovery and
deserves particular attention. This finding highlights that
englerin A and derivatives may be promising starting points
for the development of novel agents for specific therapy of
renal cell cancer by possibly avoiding common chemotherapy
related side effects.
Here, we have disclosed our second-generation approach
enabling the synthesis of an advanced englerin precursor 2 on
multigram scale. From that point, thorough SAR studies of
the two ester side chains have been conducted culminating in
the discovery of derivatives with significantly improved
activity over the natural product. Compound 16 shows
improved cytotoxicity along with increased selectivity. In
addition, we identified derivatives of the glycolate ester that
allow for the attachment of molecular handles for future
chemical biology studies. Diverting the synthesis at an earlier
stage afforded analogues with a truncated isopropyl chain,
which would be inaccessible by semisynthetic methods.
Future studies will be concerned with understanding the
molecular basis of englerinꢀs extraordinary biological profile.
Scheme 6. Structure-activity relationship (SAR) of the englerin ana-
logues. Boc=tert-butoxycarbonyl.
the tested compounds showed potent activity in these assays,
Received: December 10, 2010
it turned out that compound 16 is virtually inactive in these
Published online: March 30, 2011
Angew. Chem. Int. Ed. 2011, 50, 3998 –4002
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4001

Communications
Ishida, H. Kusama, N. Iwasawac, J. Am. Chem. Soc. 2010, 132,
8842 – 8843; d) B.-F. Sun, C.-L. Wang, R. Ding, J.-Y. Xu, G.-Q.
Lin, Tetrahedron Lett. 2010, DOI: 10.1016/j.tetlet.2010.11.087;
e) D. Parmar, K. Price, M. Spain, H. Matsubara, P. A. Bradley,
D. J. Procter, J. Am. Chem. Soc. 2011, 133, 2418 – 2420.
[8] For independent SAR studies, see: a) K. P. Chan, D. Y.-K. Chen,
ChemMedChem 2011, 6, 420 – 423; b) D. B. Ushakov, V.
Navickas, M. Strꢁbele, C. Maichle-Mꢁssmer, F. Sasse, M. E.
Maier, Org. Lett. 2011, DOI: 10.1021/ol200499t.
[9] a) A. M. Szpilman, E. M. Carreira, Angew. Chem. 2010, 122,
9786 – 9823; Angew. Chem. Int. Ed. 2010, 49, 9592 – 9628;
b) R. M. Wilson, S. J. Danishefsky, J. Org. Chem. 2006, 71,
8329 – 8351.
Keywords: guaianes · renal cancer ·
structure-activity relationships · terpenes · total synthesis
.
[1] K. A. Furge, J. P. MacKeigan B. T. Teh, Lancet Oncol. 2010, 11,
571 – 578.
[2] a) R. Ratnayake, D. Covell, T. T. Ransom, K. R. Gustafson, J. A.
Beutler, Org. Lett. 2009, 11, 57 – 60; b) J. A. Beutler, R.
Ratnayake, D. Covell, T. R. Johnson, WO 2009/088854, 2009.
[3] a) M. Willot, L. Radtke, D. Kꢁnning, R. Frꢁhlich, V. H. Gessner,
C. Strohmann, M. Christmann, Angew. Chem. 2009, 121, 9269 –
9272; Angew. Chem. Int. Ed. 2009, 48, 9105 – 9108; b) M. Willot,
M. Christmann, Nat. Chem. 2010, 2, 519 – 520.
[4] Q. Zhou, X. Chen, D. Ma, Angew. Chem. 2010, 122, 3591 – 3594;
Angew. Chem. Int. Ed. 2010, 49, 3513 – 3516.
[5] K. Molawi, N. Delpont, A. M. Echavarren, Angew. Chem. 2010,
122, 3595 – 3597; Angew. Chem. Int. Ed. 2010, 49, 3517 – 3519.
[6] K. C. Nicolaou, Q. Kang, S. Y. Ng, D. Y.-K. Chen, J. Am. Chem.
Soc. 2010, 132, 3517 – 3519.
[10] S. L. Schreiber, H. V. Meyers, K. B. Wiberg, J. Am. Chem. Soc.
1986, 108, 8274 – 8277.
[11] We show (À)-englerin B in the picture although the crystal
structure was determined for the (+)-enantiomer. CCDC 755644
((+)-englerin B) 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.
[7] a) V. Navickas, D. B. Ushakov, M. E. Maier, M. Strꢁbele, H.-J.
Meyer, Org. Lett. 2010, 12, 3418 – 3421; b) J. Xu, E. J. E. Caro-
Diaz, E. A. Theodorakis, Org. Lett. 2010, 12, 3708 – 3711; c) K.
4002 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3998 –4002