Lewis Acid Mediated Nazarov Cyclization as a Convergent and Enantioselective Entry to C-nor-D-homo-Steroids

Article abstract of DOI:10.1002/chem.201701008

A straightforward synthesis of C-nor-D-homo steroids starting from (+)-Wieland–Miescher ketone is reported. This convergent synthetic strategy utilizes a?scalable diastereoselective Nazarov cyclization of functionalized chiral aryl vinyl ketones, allowing for further functionalization. The ability to conduct this key transformation on a multi-gram scale paves the way for the synthesis of a variety of completely new C-nor-D-homo steroids, without the need of a classic steran steroid rearrangement or achiral linear reaction sequences.

Full text of DOI:10.1002/chem.201701008

DOI: 10.1002/chem.201701008
Communication
&
Steroids
Lewis Acid Mediated Nazarov Cyclization as a Convergent and
Enantioselective Entry to C-nor-D-homo-Steroids
Johannes Krieger, Toni Smeilus, Oliver Schackow, and Athanassios Giannis*[a]
bis-exo-cyclopamine (3)[8] and found them to be 10 and
Abstract: A straightforward synthesis of C-nor-D-homo
25 times more potent inhibitors of the Hh pathway than
steroids starting from (+)-Wieland–Miescher ketone is
cyclopamine (Figure 1), respectively. Importantly, both
reported. This convergent synthetic strategy utilizes a scal-
compounds proved to be stable under acidic conditions.
able diastereoselective Nazarov cyclization of function-
Much effort has been made over the past decades to devel-
alized chiral aryl vinyl ketones, allowing for further
op a valuable entry to this class of compounds.[2,9] Generally,
functionalization. The ability to conduct this key transfor-
approaches to this category of compounds can be divided into
mation on a multi-gram scale paves the way for the
two strategies. On the one hand, syntheses starting from
synthesis of a variety of completely new C-nor-D-homo
naturally or commercially available steran steroids that follow
steroids, without the need of a classic steran steroid
biomimetic routes[10] for the construction of the rearranged C-
rearrangement or achiral linear reaction sequences.
nor-D-homo skeleton, needing expensive starting materials
with limited further modification possibilities. On the other
hand are attempted total syntheses that either lead to racemic
Veratrum alkaloids belong to the group of C-nor-D-homo
steroidal alkaloids that possess the isosteroid skeleton with
a 6-6-5-6 ring pattern instead of a normal 6-6-6-5 steroid
framework.[1] The most prominent member of this family of
plant alkaloids is cyclopamine (1),[2] which inhibits the hedge-
hog signaling pathway (Hh) by binding to the heptahelical
bundle of Smoothened (Smo).[3] Therefore, inhibitors of the Hh
pathway offer new possibilities in the therapy of Hh-depen-
dent tumors.[4] Cyclopamine has further been identified as
“selective partial agonist”, capable to simultaneously inhibit
the canonical and activate the non-canonical (Smo-indepen-
dent) hedgehog signaling. Activation of the later results in
insulin-independent fast and robust glucose uptake in muscle
and adipose tissue.[5] Additionally, cyclopamine was found to
potently inhibit respiratory syncytial virus (hRSV) replication
through the noncanonical pathway, which causes severe respi-
ratory tract infections in young children as well as in elderly
and immunocompromised patients.[6] To date, no satisfactory
treatment options for hRSV-induced infections have been
reported. Hence, C-nor-D-homo steroidal alkaloids belong to
a privileged class of compounds that may have an important
impact in treatment and therapy of several diseases. However,
cyclopamine has serious drawbacks such as low potency and
chemical instability under acidic conditions. Recently, we syn-
thesized exo-cyclopamine (2) (D13-18-cyclopamine)[7] as well as
final product, or that involve long, linear reaction sequen-
ces.[10a,11] In the light of these results and the pharmacological
versatility of Veratrum alkaloids we decided to establish a gen-
eral and efficient entry to the C-nor-D-homo skeleton that also
enables the synthesis of diverse chiral C-nor-D-homo steroids
for further usage in chemical and biological investigations.
[a] J. Krieger, T. Smeilus, O. Schackow, Prof. Dr. Dr. A. Giannis
Institut fꢀr Organische Chemie
Fakultꢁt fꢀr Chemie und Mineralogie
Universitꢁt Leipzig
Johannisallee 29, 04103 Leipzig (Germany)
E-mail: giannis@uni-leipzig.de
Figure 1. A) Structures of the Hh-inhibiting C-nor-D-homo steroid cyclopa-
mine (1) and the synthetic derivatives exo-cyclopamine (2) and bis-exo-cy-
clopamine (3). B) A Nazarov cyclization disconnection to access C-nor-D-
homo steroids from simple aryl vinyl ketone precursors.
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under:
http://dx.doi.org/10.1002/chem.201701008.
Chem. Eur. J. 2017, 23, 1 – 6
1
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Scheme 1. Synthesis of the Nazarov precursors 9a–h. Reagents and conditions: a) TMSCN, KCN, 18-crown-6, toluene, RT, 16 h, 93–99%; b) LDA, 5a–h, THF,
ꢀ788C!RT, 16 h, 62–99%; c) TBAF, THF, 08C, 15 min, 81–99%; d) SOCl2, pyridine, 08C!RT, 1–3 h, 62–98%; e) THF/3m HCl (v/v 9:1), 658C, 1–3 h, 55–97%.
For this purpose we envisaged to develop a short, general
and convergent entry using the Nazarov cyclization as a key
transformation.[12] Starting from protected Wieland–Miescher
ketone (4)[13] and masked acyl anions 5a–h, derivatives 6a–h
were synthesized (62–99% yield; Scheme 1), which, after treat-
ment with tetrabutylammonium fluoride, afforded the a-
hydroxy ketones 7a–h in excellent yields (81–98%). Derivatives
5a–h were easily accessible from commercially available
aromatic aldehydes by reaction with trimethylsilylcyanide
(TMSCN) in excellent yields.[14] Attempts to use O-trimethylsilyl
Figure 2. Single-crystal X-ray structure of 10e with thermal ellipsoids scaled
to 50% probability.
cyanohydrins of benzylnitriles with ortho-alkyl substitution or
electron-withdrawing groups (-CN, -Cl, -Br, -I, -NO2) attached to
the aromatic system, completely failed to undergo nucleophilic
attack. Treatment of the a-hydroxy ketones 7a–h with SOCl2 in
pyridine afforded a,b-unsaturated ketones 8a–h (62–98%
yield). Subsequent cleavage of the acetal group proceeded
smoothly under acidic conditions, furnishing enones 9a–h in
acceptable to excellent yield (55–97%).
Having established a synthesis of 8a(H)-C-nor-D-homo
steroids, we envisaged to exploit them for the synthesis of
further highly functionalized derivatives. Inspired by the inves-
tigations of Fujioka,[15] we refined the method, enabling selec-
tive manipulation of aryl ketone moiety in presence of the A-
ring enone with the aim of adding several nucleophiles at the
C-ring carbonyl group. Exposure of 10e first to TBSOTf/PPh3 in
CH2Cl2 followed by addition of several nucleophiles afforded
compounds 11 a–f as single diastereomers in acceptable yields
(49–63%, Scheme 2). Treatment of a selection of these alcohols
with p-TsOH in refluxing toluene yielded unsaturated deriva-
tives 12a–c in good to excellent yields. Interestingly, the puta-
tive phosphonium intermediate was easily transformed to 11-
exo-methylene derivative 13 using Petasis reagent[16] (46%
yield).
Derivative 9e was chosen as model substance for the opti-
mization of the reaction conditions for the Nazarov reaction.
Treatment of this aryl vinyl ketone with a variety of Brønsted
and Lewis acids (i.e. HCl, H3PO4, TFA, CH3SO3H, BF3, AlCl3, SnCl4,
FeCl3) as well as irradiation with UV irradiation did not pro-
duced the envisaged product 10e. Gratifyingly usage of an
excess of TiCl4 afforded derivative 10e in an excellent yield as
the only reaction product. Several other C-nor-D-homo steroid
derivatives were also obtained under these conditions in multi-
gram scale in good to excellent yields (see Table 1). In the case
of compound 9b, an inseparable 2:1 mixture of regioisomers
10b-1 and 10b-2 was produced, whereas derivatives 9d and
9h failed to undergo cyclization. The stereochemistry of the
Next, derivative 10e was subjected to the classic deconjuga-
tion protocol[17] in order to shift the double bond of ring A to
ring B, a motif typical for many C-nor-D-homo steroids. Treat-
ment of 10e with AcCl/Ac2O followed by NaBH4 furnished diol
14 as single diastereomer (77% yield over two steps;
Scheme 3). Treatment of benzylic alcohol 14 with MnO2 afford-
ed aryl ketone 15, whereas subsequent benzylation of the later
1
obtained products was confirmed in all cases by H NMR spec-
troscopy, observing NOE interaction between protons H8 and
H9. These two protons did not show any NOE interaction with
the protons of the methyl group (H18). In addition, the struc-
ture of derivative 10e was confirmed by X-ray crystallography
(see Figure 2).
[18]
with Dudley reagent (benzyloxy 1-methylpyridinium triflate)
yielded derivative ether 16 in 74% yield.
O-alkylation of the enol of ketone 16 was accomplished
using Noyori’s protocol[19] and afforded enol ether 18, which
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
Table 1. Scope and limitations of the TiCl4-catalyzed Nazarov cyclization of aryl vinyl ketones 9a–h.[a]
Entry
TiCl4 [eq] Time Product
Yield [%][b] Entry TiCl4 [eq] Time Product
Yield [%][b]
9a
9.0
3.5
5.0
3 d
3 h
3 d
83
83
45
9e
9 f
9g
3.5
16 h
94
9b
15
7 d
81
74
9c
3.5
16 h
9d
15
28 d
67
9h
15
7 d
no cyclization product observed
–
[a] Substrate concentrations are 0.1m. [b] Isolated yield.
Scheme 2. Three step C11 modification of derivative 10e by in situ protec-
tion of the A-ring enone. Reagents and conditions: a) TMS- or TBSOTf, PPh3,
CH2Cl2, 08C, 1 h; b) (i) DIBAL-H or RLi or vinylmagnesium bromide, ꢀ788C to
08C; (ii) TBAF, THF, RT, 30 min; c) p-TsOH, toluene, reflux, 1 h; d) (i) Cp2TiMe2,
toluene, reflux, 2 h; (ii) TBAF, THF, RT, 30 min.
Scheme 3. Transformation of the ketone 10e to homoallylic alcohol 16. Re-
agents and conditions: a) (i) AcCl, Ac2O, 658C, 2 h, (ii) NaBH4, tBuOH/THF/
H2O, ꢀ188C!RT, 16 h, 77% over two steps; b) MnO2, CH2Cl2, RT, 16 h, 77%;
c) benzyloxy 1-methylpyridinium triflate (Dudley reagent), MgO, PhCF3, 858C,
24 h, 74%.
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
3
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
smoothly underwent Claisen rearrangement upon heating in
toluene (1308C, quantitative yield), affording 9(b)-allyl-C-nor-D-
homo steroid 19 (d.r.=2.7:1) containing a chiral quaternary C-
atom (Scheme 4). Derivative 16 was also readily transformed to
compound 17 (a C-9 epimer of 16), first by treatment with
LDA/Et2Zn/HMPA (LDA: lithium diisopropylamide, HMPA:
hexamethylphosphoramide) followed by work-up with saturat-
ed ammonium chloride solution. Indeed, the product 17 could
be isolated in 59% yield along with 29% of unchanged start-
ing material. The relative stereochemistry of 17 was unambigu-
ously established on the basis of NOE experiments showing an
interaction between protons at C-9 and C-18.
Scheme 5. Dearomatization of the D-ring by Birch reduction. Reagents and
conditions: a) Li, tBuOH, NH3, ꢀ788C, 90 min, 77%; b) Li, EtOH, NH3,
ꢀ788C!ꢀ308C, 4 h, 62%; c) HClaq, MeOH, reflux, 1 h, 67%; d) Dudley re-
agent, MgO, PhCF3, 858C, 8 h, 75%; e) (i) LDA, THF ꢀ788C, 1 h, (ii) MeI,
ꢀ658C, 4 h; f) HClaq, MeOH, reflux, 1 h, 51%.
way for synthesis and subsequent biological evaluation of a -
variety of non-natural members of the C-nor-D-homo steroidal
alkaloid class of compounds.
Scheme 4. Inversion of the C9-stereochemistry. Reagents and conditions:
a) LDA, Et2Zn, HMPA, THF, NH4Claq, ꢀ788C!RT, 48 h, 59%; b) LDA, Et2Zn,
HMPA, THF, allyliodide, ꢀ788C!RT, 48 h, 56%; e) toluene, 1308C, 8 h,
quant., d.r.=2.7:1.
Acknowledgements
This work was supported by the DFG (GI 204/8-1). We thank
Prof. Dr. H. Krautscheid and M. Sc. O. Erhart, University of
Leipzig, for the X-ray crystallography data of derivative 10e.
Moreover we thank Dr. Lothar Hennig for recording NMR
spectra and for his help in interpreting the 2D NMR spectra.
With diol 14 in our hands, the stage was set for D-ring
dearomatization. Reduction of 14 using Li/NH3/tBuOH/ꢀ788C
afforded derivative 20, whereas using Li/EtOH/NH3 at ꢀ308C
gave b,g-unsaturated ketone 21 in 62% yield (Scheme 5).
Under the influence of HCl in refluxing methanol, compound
21 underwent isomerization to steroid 22 in 67% yield. Finally,
derivative 21 was converted in two steps to compound 24
which is a C8-epimer of the steroidal part of cyclopamine.
In this work we traced the evolution of a short general and
enantioselective entry to the class of C-nor-D-homo-steroids,
which inspired and confounded chemists for almost six
decades,[10] by a convergent and highly modular approach. The
key strategic element for the formation of its tetracyclic entity
is a TiCl4-promoted diastereoselective Nazarov reaction, repre-
senting one of the few reported examples of Lewis acid medi-
ated Nazarov cyclization of aryl vinyl ketones. Highlights of
follow up modifications are the differentiation of both keto
groups by in situ tert-butylsilyl (TBS) protection, the generation
of vicinal quaternary stereocenters (C9, C10) by Claisen
rearrangement and an efficient and diastereoselective Birch
reduction/isomerization sequence yielding in C-nor-D-homo
steroid 22 as well as after methylation of the later to com-
pound 24. The approach, developed in this study, paved the
Conflict of interest
The authors declare no conflict of interest.
Keywords: electrocyclic reactions
products · steroids · umpolung
· Lewis acids · natural
[1] P. Heretsch, A. Giannis, The Alkaloids: Chemistry and Biology, Elsevier,
2015, 74, pp 201–232.
[2] a) A. Giannis, P. Heretsch, V. Sarli, A. Stçßel, Angew. Chem. Int. Ed. 2009,
48, 7911–7914; Angew. Chem. 2009, 121, 8052–8055.
[3] J. K. Chen, J. Taipale, M. K. Cooper, P. A. Beachy, Gene Dev. 2002, 16,
2743–2748.
[4] for reviews, see: a) P. Heretsch, L. Tzagkaroulaki, A. Giannis, Angew.
Chem. Int. Ed. 2010, 49, 3418–3427; Angew. Chem. 2010, 122, 3492–
3502; b) P. Heretsch, L. Tzagkaroulaki, A. Giannis, Angew. Chem. 2010,
122, 3492–3502; c) P. Heretsch, L. Tzagkaroulaki, A. Giannis, Bioorg.
Med. Chem. 2010, 18, 6613–6624; d) T. K. Rimkus, R. L. Carpenter, S.
Qasem, M. Chan, H.-W. Lo, Cancers 2016, 8, 22; e) T. N. Trinh, E. A.
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
4
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
McLaughlin, C. P. Gordon, A. McCluskey, Med. Chem. Commun. 2014, 5,
117–133.
[12] For excellent reviews, see: a) M. A. Tius, Eur. J. Org. Chem. 2005, 11,
2193–2206; b) T. N. Grant, C. J. Rieder, F. G. West, Chem. Commun. 2009,
5676–5688; c) H. Pellissier, Tetrahedron 2005, 61, 6479–6517; d) A. J.
Frontier, C. Collison, Tetrahedron 2005, 61, 7577–7606; e) D. R. Wenz, J.
Read de Alaniz, Eur. J. Org. Chem. 2015, 1, 23–37; f) S. P. Simeonov,
J. P. M. Nunes, K. Guerra, V. B. Kurteva, C. A. M. Afonso, Chem. Rev. 2016,
116, 5744–5893.
[13] A. B. Smith, T. L. Leenay, J. Am. Chem. Soc. 1989, 111, 5761–5768.
[14] K. Okazaki, K. Nomura, E. Yoshii, Synth. Commun. 1987, 17, 1021–1027.
[15] K. Yahata, M. Minami, K. Watanabe, H. Fujioka, Org. Lett. 2014, 16,
3680–3683.
[16] N. A. Petasis, E. I. Bzowej, J. Am. Chem. Soc. 1990, 112, 6392–6394.
[17] a) Z. Hua, D. A. Carcache, Y. Tian, Y.-M. Li, S.-J. Danishefsky, J. Org. Chem.
2005, 70, 9849–9856; b) T. Mase, J. Ichita, J. P. Marino, M. Koreeda, Tet-
rahedron Lett. 1989, 30, 2075–2078; c) S. S. Moon, L. M. Stuhmiller, T. C.
McMorris, J. Org. Chem. 1989, 54, 26–28; T. C. McMorris, J. Org. Chem.
1989, 54, 26–28; d) S. S. Moon, L. M. Stuhmiller, R. K. Chadha, J. C.
McMorris, Tetrahedron 1990, 46, 2287–2306; e) D. F. Taber, J. F. Berry, J.
Org. Chem. 2013, 78, 8437–8441.
[5] R. Teperino, S. Amann, M. Bayer, S. L. McGee, A. Loipetzberger, T.
Connor, C. Jaeger, B. Kammerer, L. Winter, G. Wiche, K. Dalgaard, M. Sel-
varaj, M. Gaster, R. S. Lee-Young, M. S. Febbraio, C. Knauf, P. D. Cani, F.
Aberger, J. M. Penninger, J. A. Pospisilik, H. Esterbauer, Cell 2012, 151,
414–426; M. Selvaraj, M. Gaster, R. S. Lee-Young, M. S. Febbraio, C.
Knauf, P. D. Cani, F. Aberger, J. M. Penninger, J. A. Pospisilik, H. Esterba-
uer, Cell 2012, 151, 414–426.
[6] B. Bailly, C.-A. Richard, G. Sharma, L. Wang, L. Johansen, J. Cao, V. Pend-
harkar, D.-C. Sharma, M. Galloux, Y. Wang, R. Cui, G. Zou, P. Guillon, M.
von Itzstein, J.-F. Elꢁouꢂt, R. Altmeyer, Sci. Rep. 2016, 6, 25806.
[7] P. Heretsch, A. Bꢃttner, L. Tzagkaroulaki, S. Zahn, B. Kirchner, A. Giannis,
Chem. Commun. 2011, 47, 7362–7364.
[8] J. Moschner, A. Chentsova, N. Eilert, I. Rovardi, P. Heretsch, A. Giannis,
Beilstein J. Org. Chem. 2013, 9, 2328–2335.
[9] S. Gao, Q. Wang, C. Chen, J. Am. Chem. Soc. 2009, 131, 1410–1412.
[10] for an overview, see: a) S. Gao, C. Chen in Total synthesis of natural prod-
ucts: At the frontiers of organic chemistry (Eds.: J. J. Li, E. J. Corey);
Springer: Berlin, London, 2012, pp. 25–37; b) P. Heretsch, S. Rabe, A.
Giannis, J. Am. Chem. Soc. 2010, 132, 9968–9969; c) X. Fu, T.-M. Chan,
C.-H. Tann, T. K. Thiruvengadam, Steroids 2002, 67, 549–554.
[11] a) S. D. Guile, J. E. Saxton, M. Thornton-Pett, J. Chem. Soc. Perkin Trans.
1 1992, 1, 1763–1767; b) E. Brown, J. Lebreton, Tetrahedron 1987, 43,
5827–5840.
[18] J. Tummatorn, P. A. Albiniak, G. B. Dudley, J. Org. Chem. 2007, 72, 8962–
8964.
[19] Y. Morita, M. Suzuki, R. Noyori, J. Org. Chem. 1989, 54, 1785–1787.
Manuscript received: March 6, 2017
Final Article published: && &&, 0000
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
COMMUNICATION
&
Steroids
J. Krieger, T. Smeilus, O. Schackow,
A. Giannis*
&& – &&
Lewis Acid Mediated Nazarov
Cyclization as a Convergent and
Enantioselective Entry to C-nor-D-
homo-Steroids
Titan strength: A TiCl4-mediated
Nazarov cyclization of chiral aryl vinyl
ketones enabled a convergent and
enantioselective five-step synthesis
towards the class of C-nor-D-homo
steroids.
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!