Enantioselective hydroarylation of bridged [3.2.1] heterocycles: An efficient entry into the homoepibatidine skeleton

Article abstract of DOI:10.1021/ol401477r

Achiral [3.2.1] bridged heterocycles containing a bridging amide can undergo enantioselective hydroarylation reactions under rhodium(I) catalysis. These reactions proceed in high yield and enantioselectivity in most cases, under mild reaction conditions and using commercially available Josiphos ligands. The phosphine ligand structure and the protecting group on the nitrogen both have significant effects on the selectivity and yield of the reactions.

Full text of DOI:10.1021/ol401477r

ORGANIC
LETTERS
2013
Vol. 15, No. 13
3424–3427
Enantioselective Hydroarylation of
Bridged [3.2.1] Heterocycles: An Efficient
Entry into the Homoepibatidine Skeleton
~
Ryan A. Brawn,* Cristiano R. W. Guimaraes, Kim F. McClure, and Spiros Liras
CVMED Chemistry, Pfizer Worldwide Research and Development,
445 Eastern Point Road, Groton, Connecticut 06340, United States
ryan.brawn@pfizer.com
Received May 24, 2013
ABSTRACT
Achiral [3.2.1] bridged heterocycles containing a bridging amide can undergo enantioselective hydroarylation reactions under rhodium(I)
catalysis. These reactions proceed in high yield and enantioselectivity in most cases, under mild reaction conditions and using commercially
available Josiphos ligands. The phosphine ligand structure and the protecting group on the nitrogen both have significant effects on the
selectivity and yield of the reactions.
Enantioselective reactions that desymmetrize achiral
starting materials are efficient means of forming small
molecules with one or more chiral centers.1 In particular,
the enantioselective functionalization of meso-bridged het-
erocyclic systems provides a rapid entry into chiral building
blocks with well-defined conformations which have medic-
inal chemistry applications, of relevance to the pursuit of
higher potency and target selectivity.2 Despite the utility of
these compounds, efficient reactions to form and selectively
functionalize such building blocks remain underdeveloped.
One reaction that can be used to efficiently desymmeter-
ize bridged heterocycles is the enantioselective hydro-
arylation of alkenes.3 The Lautens group has reported
enantioselective hydroarylation reactions of bridged het-
erocycles that form the desired products in moderate
to high yield and enantioselectivity, typically starting from
bicyclic hydrazines.4 One limitation to this chemistry is
the lack of examples with a bridging heteroatom, as these
examples typically result in the opening of the high energy
bridge system.5 We have recently reported an efficient
synthesis of meso [3.2.1] bridged heterocycles through a
(1) For some recent reviews, see: (a) Li, Z.; Yamamoto, H. Acc.
Chem. Res. 2013, 46, 506–518. (b) Enriquez-Garcia, A.; Kuendig, E. P.
Chem. Soc. Rev. 2012, 41, 7803–7831. (c) Hennecke, U. Angew. Chem.,
Int. Ed. 2012, 51, 4532–4534. (d) Diaz de Villegas, M. D.; Galvez, J. A.;
Etayo, P.; Badorrey, R.; Lopez-Ram-de-Viu, P. Chem. Soc. Rev. 2011,
40, 5564–5587. (e) Zhou, Z.; Wang, Y.; Tang, C. Curr. Org. Chem. 2011,
15, 4083–4107. (f) Woodmansee, D.; Pfaltz, A. Chem. Commun. 2011,
47, 7912–7916.
(2) (a) Zask, A.; Kaplan, J.; Verheijean, J. C.; Richard, D. J.; Curran,
K.; Brooijmans, N.; Bennett, F. M.; Toral-Barza, L.; Hollander, I.;
Ayral-Kaloustian, S.; Yu, K. J. Med. Chem. 2009, 52, 7942–7945. (b)
Curran, K. J.; Verheijean, J. C.; Kaplan, J.; Richard, D. J.; Toral-Barza,
L.; Hollander, I.; Lucas, J.; Ayral-Kaloustian, S.; Yu, K.; Zask, A.
Bioorg. Med. Chem. Lett. 2010, 20, 1440–1444. (c) Li, H.; Xu, R.; Cole,
D.; Clader, J. W.; Greenlee, W. J.; Nomeir, A. A.; Song, L.; Zhang, L.
Bioorg. Med. Chem. Lett. 2010, 20, 6606–6609. (d) Zask, A.; Verheijen,
J. C.; Richard, D. J.; Kaplan, J.; Curran, K.; Toral-Barza, L.; Lucas, J.;
Hollander, I.; Yu, K. Bioorg. Med. Chem. Lett. 2010, 20, 2644–2647.
(3) (a) Yang, L.; Guo, X.; Li, C.-J. Adv. Synth. Catal. 2010, 352,
2899–2904. (b) Ito, S.; Itoh, T.; Nakamura, M. Angew. Chem., Int. Ed.
2011, 50, 454–457. (c) Shao, C.; Yu, H.-J.; Wu, N.-Y.; Tian, P.; Wang,
R.; Feng, C.-G.; Lin, G.-Q. Org. Lett. 2011, 13, 788–791. (d) Sevov,
C. S.; Hartwig, J. F. J. Am. Chem. Soc. 2013, 135, 2116–2119.
(4) (a) Panteleev, J.; Menard, F.; Lautens, M. Adv. Synth. Catal.
2008, 350, 2893–2902. (b) Menard, F.; Lautens, M. Angew. Chem., Int.
Ed. 2008, 47, 2085–2088. (c) Bexrud, J.; Lautens, M. Org. Lett. 2010, 12,
3160–3163.
(5) (a) Lautens, M.; Hiebert, S.; Renaud, J.-L. J. Am. Chem. Soc.
2001, 123, 6834–6839. (b) Lautens, M.; Dockendorff, C.; Fagnou, K.;
Malicki, A. Org. Lett. 2002, 4, 1311–1314. (c) Lautens, M.; Dockendorff,
C. Org. Lett. 2003, 5, 3695–3698. (d) Cheng, H.; Yang, D. J. Org. Chem.
2012, 77, 9756–9765. (e) Tsui, G. C.; Tsoung, J.; Dougan, P.; Lautens,
M. Org. Lett. 2012, 14, 5542–5545.
~
(6) Brawn, R. A.; Guimaraes, C. R. W.; McClure, K. F.; Liras, S.
Org. Lett. 2012, 14, 4802–4805.
r
10.1021/ol401477r
Published on Web 06/21/2013
2013 American Chemical Society

bis-amination/ring closing metathesis sequence.6 Herein
we report an efficient, Rh(I) catalyzed enantioselective
hydroarylation of these heterocycles using commercially
available ligands under mild reaction conditions. The
hydroarylation products are stable to the reaction con-
ditions, and no ring opening is observed (Scheme 1).
One application of this methodology is the synthesis of
previously unreported analogues of homoepibatidine
(Scheme 1, box), a known nicotinic acetylcholine recep-
tor (nAChR) ligand.
Table 1. Optimization of Hydroarylation Conditions
yielda
(%)
eeb
(%)
R
ligand
solvent
temp
productc
Cbz
L1
L2
L3
L4
L1
L2
L3
L5
L1
L1
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
80
84
2a
2a
2a
2a
2b
2b
2b
2b
2c
Scheme 1. Hydroarylations of [3.2.1] Bridged Heterocycles
Cbz
38
78
Cbz
0
N/A
30
Cbz
47
Boc
Boc
76
17
95
78
Boc
trace
trace
83
NA
NA
78
Boc
CO2Me
cumyl
0
N/A
N/A
a Yields refer to isolated products after purification over silica. b Ee
determined by chiral HPLC. c All reactions run with 5 mol % rhodium,
10 mol % ligand, 2.0 equiv of NEt3, and 1.5 equiv of boronic acid at rt
for 16 h.
The ligands initially screened for the asymmetric hydro-
arylation reactions were based on the Josiphos ferrocene/
diphosphine complex (Table 1, box), previously used by
the Lautens group.4 These ligands are convenient not only
because of their previously reported reactivity but also
because both enantiomers and a number of different
phosphines are commercially available.7 Other phosphine
ligand classes (Walphos, DIOP, BINAP, etc.) typically
gave poor conversions to the desired products, even at
elevated temperatures. After extensive reaction screening it
was found that the phenyl/tert-butyl Josiphos ligand (L1)
gave the highest yields of the desired hydroarylation
product. No ring-opening products were observed under
any of the reaction conditions screened.
The reactions were optimized using phenylboronic acid
along with the rhodium(I) source and ligand (Table 1).
Interestingly, the nitrogen protecting group had a major
impact on the yield and enantioselectivity of the hydro-
arylations. The cumyl protecting group previously re-
ported gave rise to modest conversion under a variety of
reaction conditions, including at elevated temperatures.
Neutralizing the basicity of the nitrogen, by switching to a
carbamate, allowed clean conversion to the desired pro-
duct in high isolated yield under mild reaction conditions.8
The steric bulk of the carbamate also had a significant
impact on enantioselectivity, with the bulky Boc group
leading to the product with the highest ee. Clearly the steric
bulk on thebridgingnitrogen has aneffect on theapproach
of the catalyst to the olefin.
group on the phosphine next to the chiral center (PR02)
was necessary for high enantioselectivity, suggesting that
this center is key to the facial selectivity of the nucleophilic
addition. In all cases a single diastereomer (exo) was
observed. The base used in the reaction had little effect,
as other mild bases (Hunig’s base, K2CO3) gave similar
results as triethylamine.
These observations could be further explained by
in silico modeling of the reaction transition state. The
starting configuration for the Josiphos ligand L1, the
one that provided the best yields and enantioselectivities,
was built using a small-molecule crystal structure obtained
from the Cambridge Structural Database (CSD) (refcode =
CAQSAP). The ligand in CSD has R = cyclohexyl and
R0 = tert-butyl. In that structure, the rhodium ion is
coordinated with the phenyl anion and iodide, in addition
to the two phosphine groups. To build models for the
transition states (TSs) that lead to the observed stereo-
chemistry using the Josiphos ligand L1 as the catalyst,
the cyclohexyl groups were changed to phenyl and the
iodide ion was deleted. The bridged heterocycle protected
by the Boc group was positioned in two vectors, approach-
ing the reactioncenter adjacently tothe phenyl (TS1) or the
tert-butyl groups (TS2) (Figure 1). The phenyl anion was
built in the remaining vector of the rhodium(I) square
planar coordination geometry in each case. The bridged
The phosphines on the Josiphos ligands also had a
significant impact on the yield and ee of the reactions,
again due to steric factors. Higher steric bulk on the
phosphine (PR2) gave lower yields. A bulky tert-butyl
(7) The ligands used in this paper were purchased from Aldrich and
Strem.
(8) See Supporting Information for details on formation of the
carbamate protected bridged heterocycles.
Org. Lett., Vol. 15, No. 13, 2013
3425

Table 2. Hydroarylation Reactions
entry
Ar
yielda
eeb
productc
1
Ph
76
89d
73
85d
81
82
87
82
69d
36d
55e
17e
83
0
95
90
92
90
89
93
96
88
94
50
42
76
77
N/A
N/A
2b
3a
3b
3c
3d
3e
3f
2
p-CF3Ph
p-BrPh
3
4
p-CO2MePh
p-Tol
5
6
p-NHAcPh
m-OMePh
m-BrPh
7
8
3g
3h
3i
9
m-CO2MePh
o-FPh
10
11
12
13
14
15
Figure 1. Models for the TS structures that lead to the observed
stereochemistry.
o-FPh
3i
o-Tol
3j
3-thiophene
m-NO2Ph
o-COoMePh
3k
N/A
N/A
heterocycle double bond was oriented in a way to provide
the observed stereochemistry when the newly CꢀC bond
with the phenyl anion is formed. Two TS models were
obtained (Figure 1). The energy difference at the B3LYP/
LACV3P**þ //B3LYP/LACV3P* level between the two
TS models is ꢀ0.32 kcal/mol favoring the one where the
bridged heterocycle approaches the reaction center adja-
cently to the phenyl groups (cyan). Addition of zero-point
vibrational energies and thermal contributions to enthalpy
at 25 °C increased the difference to ꢀ1.60 kcal/mol.
Addition of entropic contributions led to a reduced free-
energy energy difference in the gas phase of ꢀ0.29 kcal/
mol. Incorporation of THF solvent effects provided a final
value of ꢀ0.79 kcal/mol. The TS1 model, the more stable
transition state, helps explain some of the results listed in
Table 1. For example, theelectron-withdrawingCF3 group
at the 4-position of the phenyl rings (L2) would reduce the
electron density of the phenyl anion, negatively impacting
the carborhodation step. Higher steric bulk on one phos-
phine (R) in L3 makes the catalyst too crowded for the
bridged heterocycle to approach the reaction center. In the
case of L5, the model suggests that the addition of several
substituents to the phenyl groups would lead to clashes
with the protecting groups.
The hydroarylation reaction worked for a variety of aryl
boronic acids, giving high yields and enantioselectivities in
most cases. Para- and meta-substituted boronic acids typi-
cally gave high yields, although cases with an electron-
withdrawing group necessitate the addition of extra equiva-
lents of boronic acid to push the reaction to completion
(Table 2, entries 2ꢀ9). All of these examples gave high
enantioselectivity using the commercially available Josiphos
ligand L1. The only functional group issue observed was
with the strongly withdrawing meta-nitro substituent, which
gave no reaction, even with extra equivalents of the boronic
acid and elevated temperatures. The model in Figure 1 also
0
a Yields refer to isolated products after purification over silica. b Ee
determined by chiral HPLC. c All reactions run with 5 mol % rhodium,
10 mol % ligand, 2.0 equiv of NEt3, and 1.25 equiv of boronic acid at rt
for 16 h. d Reaction run with 2.5 equiv of boronic acid. e Reaction run at
50 °C with 2.5 equiv of boronic acid.
explains the reduced yields obtained with the ortho-substi-
tuted phenyl boronic acids. These groups would lead to
rotation of the phenyl anion and consequently worse orbital
overlap with the bridged heterocycle double bond carbon in
the transition state. The ortho-fluorophenyl boronic acid
gave a moderate yield and low enantioselectivity at rt and
only a slightly higher yield when the temperature was raised.
The ortho-tolylboronic acid gave a low yield and the ortho-
methyl ester gave no reaction, showing a clear trend based
on the size of the substituent at the ortho position. Heating
the reaction helped the yields slightly, but resulted in slightly
lower levels of enantioselectivity. One heterocycle, 3-thio-
phene boronic acid, gave the desired product in high yield
and moderate enantioselectivity (3k, Table 2).
Having demonstrated the incorporation of heterocycles
in suchhydroarylations, we decided toapplythischemistry
to a previously unreported oxygen-containing analogue of
homoepibatidine.9 Reaction of 1b with 4-chloro-3-pyri-
dine boronic acid gave the desired product (ꢀ)-4 in mod-
erate yield and selectivity upon heating (Scheme 2). The
absolute stereochemistry of the products was assigned by
X-ray crystallography of two antipodes of hydroarylation
product 4 (Figure 2).10
(9) (a) Xu, R.; Bai, D.; Chu, G.; Tao, J.; Zhu, X. Bioorg. Med. Chem.
Lett. 1996, 6, 279–282. (b) Bai, D.; Xu, R.; Chu, G.; Zhu, X. J. Org.
Chem. 1996, 61, 4600–4606. (c) Malpass, J. R.; Hemmings, D. A.; Wallis,
A. L.; Fletcher, S. R.; Patel, S. J. Chem. Soc., Perkin Trans. 1 2001, 1044–
1050.
(10) See Supporting Information for detailed X-ray analysis.
3426
Org. Lett., Vol. 15, No. 13, 2013

Scheme 2. Synthesis of Homoepibatidine Analogue 4a
Scheme 3. Synthesis of Homoepibatidine Analogue 5a
a Ee determined by chiral HPLC. Reaction run with 5 mol %
rhodium, 10 mol % ligand, 2.0 equiv of NEt3 at 50 °C with 2.5 equiv
of boronic acid.
in an R4β2 nAChR assay.11 Interestingly, 5a showed
minimal binding to several other receptors, including the
muscarinic acetylcholine receptor mAChR M1.12
In conclusion an enantioselective hydroarylation of
heterobicyclic olefins has been achieved, using examples
containing a bridging nitrogen. The majority of the
reactions proceeded in high yield and enantioselectivity
using commercially available Josiphos ligands under
mild conditions. No ring-opening byproducts were ob-
served in any of the cases screened. Modeling work
helped explain the stereochemical course of the reactions
as well as the effects of different ligands and protecting
groups on nitrogen. One of the products formed is a
previously unreported analogue of the biologically active
product homoepibatidine.
Acknowledgment. R.A.B. would like to thank Pfizer for
the postdoctoral fellowship and Dr. Allyn Londregan and
Dr. Adam Kamlet (Pfizer CVMED) for helpful discus-
sions. The authors would like to thank Dr. Brian Samas
(Pfizer) for obtaining X-ray data of 4, James Bradow
(Pfizer) for help with HPLC, Kelly McKiernan (Pfizer)
for running biological assays, and Dr. Dennis Anderson
(Pfizer) for help with NMR.
Figure 2. X-ray structure of (ꢀ)-4.
Deprotection of the Boc group in 4 led to the desired
homoepibatidine analogue 5a (Scheme 3). Compound 5a
showed dose dependent inhibition of the flow of current
resulting fron 1 μM acetylcholine, with an IC50 of 0.17 μm
Supporting Information Available. Preparation and
characterization of all new compounds is included. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(11) (a) Kim, Y.-H.; Oh, C.-Y.; Lee, K.-Y.; Lee, Y.-S.; Jung, Y.-H.;
Park, H.-J.; Ham, W.-H. Arch. Pharm. Res. 2002, 25, 49–52. (b) Bremner,
J. B.; Godfrey, C. A.; Jensen, A. A.; Smith, R. J. Bioorg. Med. Chem. Lett.
2004, 14, 271–273. (c) Carroll, F. I. Bioorg. Med. Chem. Lett. 2004, 14,
1889–1896. (d) White, R.; Malpass, J. R.; Handa, S.; Baker, S. R.; Broad,
L. M.; Folly, L.; Mogg, A. Bioorg. Med. Chem. Lett. 2006, 16, 5493–5497.
(e) Mu, L.; Drandarov, K.; Bisson, W. H.; Schibig, A.; Wirz, C.; Schubiger,
P. A.; Westera, G. Eur. J. Med. Chem. 2006, 41, 640–650.
(12) See Supporting Information for a complete table of receptors
and binding results.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 13, 2013
3427