Enantioselective Fluorination of Spirocyclic β-Prolinals Using Enamine Catalysis

Article abstract of DOI:10.1055/s-0036-1588100

A series of spirocyclic carbaldehydes were successfully fluorinated using enamine catalysis, furnishing the corresponding tertiary fluorides in both high yields and enantioselectivities. The fluorinated spirocycles provide a set of novel building blocks interesting from a medicinal chemistry point of view.

Full text of DOI:10.1055/s-0036-1588100

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2016, 27, A–D
letter
A
K. Fjelbye et al.
Letter
Synlett
Enantioselective Fluorination of Spirocyclic β-Prolinals Using
Enamine Catalysis
Kasper Fjelbyeb
Mauro Marigoa
Rasmus Prætorius Clausenb
Karsten Juhl*a
8 examples
♦ up to 97% ee
♦ up to 91% yield
♦ low catalyst loading
♦ spirocyclic building
blocks
OTMS
Ar
Ar
X
X
CHO
+
CHO
F
n
n
N
H
n
n
NFSI
Ar = 3,5-(F3C)2C6H3
N
N
Pg
Pg
a Discovery Chemistry & DMPK, H. Lundbeck A/S, Ottiliavej 9, 2500 Valby,
Copenhagen, Denmark
n = 1, 2
X = O, N-Boc
Pg = Boc, Cbz
KAJU@Lundbeck.com
b Department of Drug Design and Pharmacology, Faculty of Health and
Medical Sciences, University of Copenhagen, 2 Universitetsparken,
2100 Copenhagen, Denmark
Received: 17.08.2016
Accepted after revision: 26.10.2016
Published online: 15.11.2016
enamine catalysis, a subject that has been thoroughly stud-
ied for nonbranched aldehydes a decade ago.4 However,
only recently were a series of simple acyclic α-branched al-
dehydes fluorinated with high enantioselectivity employ-
ing primary amines as organocatalysts.5 Based on the well-
established methodology for α-monosubstituted aldehydes
using a commercially available fluorine source, N-fluoro-
benzenesulfonimide (NFSI), the fluorination of the spirocy-
clic compound 1a comprised of an oxetane and a pyrrodine
was examined. In these reactions a series of structurally
different organocatalysts in methyl tert-butyl ether (MTBE)
or a mixture of THF and i-PrOH at 40 °C were tested
(Scheme 1).4a,b,6 The reaction catalyzed by proline furnished
the fluorinated aldehyde 2a in high yield, although with
moderate control of the stereochemical outcome. The use of
three branched chiral primary amines as catalysts fur-
nished the opposite enantiomer of 2a, albeit with poor se-
lectivity. The use of imidazolidinone catalyst 7 showed an
increased enantioselectivity relative to proline and provid-
ed 2a in 76% yield. Encouragingly, well-known silylated dia-
ryl prolinol 9 catalyzed the fluorination of 1a in both high
yield and with 79% ee. In order to further establish the sub-
strate scope of the enantioselective fluorination using 9 on
spirocyclic frameworks, a range of pyrrolidine-based sub-
strates 1a–h were prepared using either a 1,3-dipolar cy-
cloaddition protocol or an intramolecular cyclization ap-
proach.7
DOI: 10.1055/s-0036-1588100; Art ID: st-2016-b0535-l
Abstract A series of spirocyclic carbaldehydes were successfully fluori-
nated using enamine catalysis, furnishing the corresponding tertiary
fluorides in both high yields and enantioselectivities. The fluorinated
spirocycles provide a set of novel building blocks interesting from a me-
dicinal chemistry point of view.
Key words spirocycles, organocatalysis, enantioselective catalysis,
fluorination, asymmetric synthesis
In modern drug discovery, the ability to tailor the prop-
erties of lead compounds by selective introduction of fluo-
rine, affecting parameters such as permeability, pKa, and
potency, has provided medicinal chemists with a useful tool
for a more innovative design of active pharmaceutical in-
gredients.1 Despite the potentially rewarding impact of in-
troducing fluorine, accessing more advanced fluorinated
scaffolds is mainly limited by a lack of synthetic methodolo-
gy.2 Another important parameter in medicinal chemistry
research is the synthetic accessibility of spatially diverse
building blocks with a defined three-dimensional geome-
try. The control of molecular geometry provides the possi-
bility for better vectorization, allowing functional groups to
be directed towards regions that are optimal for target in-
teraction.3 Spirocyclic building blocks constitute a structur-
ally attractive compound class, in many ways similar to reg-
ular saturated heterocycles. However, the conformational
constraints of the spirocyclic structures allow access to dif-
ferent chemical space relative to the native heterocycles,
thus providing medicinal chemists with increased building-
block diversity. Herein, we present the enantioselective fluo-
rination of spirocyclic carbaldehydes to yield a range of nov-
el fluorinated building blocks, successfully facilitated using
The scope of the enantioselective fluorination of spiro-
cyclic carbaldehydes was initially examined by studying the
tolerability of different carbamate-based N-protecting
groups by comparing the enantioselectivity in the fluorina-
tion of 1a and 1b (Scheme 2). As the fluorination of the N-
Boc-protected analogue 1b furnished the corresponding
product 2b in high yield with 77% ee, an outcome similar to
what was obtained for 2a, the N-protecting group seemed
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

B
K. Fjelbye et al.
Letter
Synlett
O
O
CHO
F
R1
R1
CHO
NFSI (1.2 equiv)
catalyst (10 mol%)
CHO
CHO
NFSI (1.2 equiv)
9 (10 mol%)
R1
R1
F
R2
R2
R2
solvent, 40 °C, 18 h
MTBE, 40 °C, 18 h
R2
N
N
N
N
Cbz
1a
Cbz
2a
Pg
1a–h
Pg
2a–h
O
O
F
CHO
CHO
COOH
CHO
N
F
F
H
NH2
NH2
O
N
N
N
3
4
5
Cbz
Cbz
Boc
99% yield
51% ee
8% yield
–14% ee
52% yield
–12% ee
2a
2b
2c
71% yield
79% ee
77% yield
77% ee
91% yield
79% ee
O
O
N
H2N
Boc
• 3HCl
OMe
N
N
• DCA
N
F
CHO
CHO
Bn
Bn
tBu
N
N
H
CHO
F
F
H
N
N
N
N
6
7
8
Boc
Boc
Cbz
44% yield
–4% ee
76% yield
42% yield
48% ee
62% ee
2d
2e
2f
68% yield
26% ee
66% yield
97% ee
68% yield
78% ee
OTMS
Ar
Ar
N
Boc
O
CHO
N
CHO
H
F
F
9
73% yield
N
N
79% ee
Boc
Boc
Scheme 1 Reactions performed on 0.1 mmol scale. Solvent = MTBE
(0.2 M) for 4, 5 and 9, whereas i-PrOH–THF (1:9, 0.2 M) was used for 3,
6–8. The yields reported are isolated yields. The experiment with 9 was
also conducted using 1 mol% catalyst with no loss in ee, albeit with sig-
nificantly decreased conversion after 18 h. Ar = 3,5-(F3C)2C6H3,
DCA = 2,2-dichloroacetic acid. The ee was determined using chiral SFC
after reduction to the corresponding alcohol.
2g
57% yield
93% ee
2h
67% yield
87% ee
Scheme 2 The ee was determined using chiral SFC after reduction to
the corresponding alcohol or on a p-nitrobenzoyl derivative; see the
Supporting Information for further details. The yields reported corre-
spond to the isolated yields of the fluorinated aldehydes.
to have little influence on the enantioselectivity. However,
when an N-benzyl protecting group was used, the reaction
yielded only trace amounts of the desired product.
pyrrolidine or azetidine unit after chemoselective N-depro-
tection. The fluorination protocol using 9 as catalyst also
tolerated six-membered rings spiro-fused at the 4-position
of the pyrrolidine, including a tetrahydropyranyl (2g) and
an N-Boc-protected piperidine (2h), furnishing the fluori-
nated products in high yields and enantioselectivities.
The absolute configuration of 2e was confirmed using
X-ray crystallography for 10, a crystalline derivative of 2e,
being the product formed in highest ee (Figure 1). Further-
more, the absolute configuration of 2e was also confirmed
by vibrational circular dichroism (VCD) analysis of the cor-
responding alcohol.8
A similar level of enantioselectivity was observed for 1c,
where the spirooxetane is attached to the other side of the
aldehyde on the pyrrolidine scaffold relative to 1a. In the
case of 2d, without branching at the β-positions, an expect-
ed loss of enantioselectivity was observed. We propose that
this result is a consequence of lower differentiation of the
two α-substituents leading to a mixture of E- and Z-enam-
ine intermediates, which give different stereochemical out-
come, and not due to insufficient face shielding of the cata-
lyst. When the pyrrolidine was substituted with gem-di-
methyl substituents in place of the oxetane ring, as was the
case for 1e, the enantioselectivity drastically increased to
97% ee in the formation of 2e, presumably due to the in-
creased steric demand of the gem-dimethyl groups, and
thus better control of the enamine E/Z ratio, relatively to the
more strained oxetane ring. The substrate scope was fur-
ther extended to include a spiroazetidine moiety (1f), bear-
ing a N-Cbz group, and thus orthogonally protected to the
pyrrolidine nitrogen. This provides an opportunity for di-
versification upon further functionalization of either the
Figure 1 ORTEP diagram of 10 showing two molecules of 10 in the
asymmetric unit cell. The crystals were obtained from boiling EtOH
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

C
K. Fjelbye et al.
Letter
Synlett
The reaction mechanisms of stereoselective reactions of
aldehydes catalyzed by secondary amines have been stud-
ied previously.9 Based on these studies, we rationalize the
stereochemical outcome imposed by catalyst 9 by the sum
of two effects: the pyrrolidine substituents’ impact on the
enamine’s E/Z equilibrium, resulting in a distinct enamine
configuration, and the face-shielding ability of the catalyst,
ultimately controlling the asymmetric induction (Scheme
3). Thus, a high enamine E/Z equilibrium constant com-
bined with a fast equilibration rate relative to the rate of
fluorination should allow for a high control of the enantio-
selectivity.10
Acknowledgment
The work was financially supported by The Danish Agency for Sci-
ence, Technology and Innovation and H. Lundbeck A/S. We thank Jes-
sica Giacoboni MSc (H. Lundbeck A/S) for her help during compound
analysis, Henrik Pedersen PhD (H. Lundbeck A/S) for support of the
SFC instrument used for determination of optical purity. Further-
more, a special thanks to The European Centre for Chirality (EC2) and
BioTools for determination of absolute configuration.
Supporting Information
Supporting information for this article is available online at
http://dx.doi.org/10.1055/s-0036-1588100.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
o
nrtogI
f
rmoaitn
The enamine's re face is
shielded by the catalyst
a) 4-spiro compounds
Ar
References and Notes
OTMS
(R)
Ar
(1) (a) Huchet, Q. A.; Kuhn, B.; Wagner, B.; Kratochwil, N. A.;
Fischer, H.; Kansy, M.; Zimmerli, D.; Carreira, E. M.; Müller, K.
J. Med. Chem. 2015, 58, 9041. (b) Purser, S.; Moore, P. R.;
Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37, 320.
(2) (a) Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
(b) Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.;
Meanwell, N. A. J. Med. Chem. 2015, 58, 8315.
O
O
CHO
F
N
(Z)
NFSI
H2O
+ catalyst
N
N
Cbz
Cbz
The enamine's si face is
exposed to NFSI
(3) Carreira, E. M.; Fessard, T. C. Chem. Rev. 2014, 114, 8257.
(4) (a) Marigo, M.; Fielenbach, D.; Braunton, A.; Kjærsgaard, A.;
Jørgensen, K. A. Angew. Chem. Int. Ed. 2005, 44, 3703. (b) Beeson,
T. D.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127, 8826.
(c) Steiner, D. D.; Mase, N.; Barbas, C. F. Angew. Chem. Int. Ed.
2005, 44, 3706. (d) Enders, D.; Huttl, M. R. M. Synlett 2005, 991.
The enamine's re face is
exposed to NFSI
b) 2-spiro compounds
Ar
(R)
Ar
OTMS
N
(E)
F
CHO
NFSI
H2O
For
a
comprehensive review on amine catalysis, see:
+ catalyst
(e) Melchiorre, P.; Marigo, M.; Carlone, A.; Bartoli, G. Angew.
Chem. Int. Ed. 2008, 47, 6138.
O
N
O
N
Cbz
Cbz
(5) For initial fluorinations of α-branched aldehydes with up to 90%
ee and moderate yields, see: (a) Brandes, S.; Niess, B.; Bella, M.;
Prieto, A.; Overgaard, J.; Jørgensen, K. A. Chem. Eur. J. 2006, 12,
6039. (b) Witten, M. R.; Jacobsen, E. N. Org. Lett. 2015, 17, 2772.
(c) Shibatomi, K.; Kitahara, K.; Okimi, T.; Abe, Y.; Iwasa, S. Chem.
Sci. 2016, 7, 1388.
The enamine's si face is
shielded by the catalyst
Scheme 3 Enamine transition-state geometries used to rationalize the
observed stereochemical outcome of the fluorination reactions
(6) For recent findings on organocatalytic fluorination of aldehydes
bearing two chiral centers, see: (a) Fjelbye, K.; Marigo, M.;
Clausen, R. P.; Juhl, K. Org. Lett. 2016, 18, 1170. (b) Hu, X.; Lawer,
A.; Peterson, M. B.; Iranmanesh, H.; Ball, G. E.; Hunter, L. Org.
Lett. 2016, 18, 662.
(7) For the synthesis of the spirocyclic structures, see: Fjelbye, K.;
Marigo, M.; Clausen, R. P.; Juhl, K. Synlett 2016, DOI: 10.1055/s-
0036-1588902.
(8) For further details on the determination of absolute stereo-
chemistry, see the Supporting Information.
(9) (a) Bahmanyar, S.; Houk, K. N. J. Am. Chem. Soc. 2001, 123,
11273. (b) Mangion, I. K.; Northrup, A. B.; MacMillan, D. W. C.
Angew. Chem. Int. Ed. 2004, 43, 6722.
For the substrates spiro-fused at the 4-position (Scheme
3, a), the Z-enamine is presumed to be dominant having the
pyrrolidine ring of the catalyst placed farthest away from
the oxetane, and the steric bulk of the catalyst away from
the N-Cbz group, leaving the si face exposed to fluorination.
In contrast, the re face is presumed to be the most accessi-
ble site of fluorination for the substrates substituted at the
2-position due to a predominant presence of the E-enamine
based on similar rationalization of minimizing steric inter-
actions (Scheme 3, b).
In conclusion; a series of fluorinated spirocyclic carbal-
dehydes were successfully prepared in high yields and en-
antioselectivities using enamine catalysis, demonstrating
the synthetic accessibility of such building blocks interest-
ing in particular from a medicinal chemistry point of
view.11
(10) For a mechanistic study of enamine catalysis on α-branched
aldehydes, see: Bures, J.; Armstrong, A.; Blackmond, D. G. Chem.
Sci. 2012, 3, 1273.
(11) General Procedure for the Enantioselective Fluorination of
Aldehydes
To a solution of 1a–h in MTBE (0.5 M) was added catalyst (R)-9
(10 mol%), and the mixture was stirred for 5 min at r.t. after
which N-fluorobenzenesulfonimide (NFSI, 1.2 equiv) was
added, and the resulting mixture was stirred at 40 °C for 18 h.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

D
K. Fjelbye et al.
Letter
Synlett
Subsequently, the reaction mixture was diluted with MTBE and
passed through a filter directly onto SiO2 for flash chromato-
graphic purification to yield the desired product.
Benzyl (S)-8-Fluoro-8-formyl-2-oxa-6-azaspiro[3.4]octane-
6-carboxylate (2a)
The enantioselective fluorination procedure provided 2a (76
mg, 0.259 mmol, 71% yield, 79% ee) as a colorless oil. 1H NMR
(600 MHz, CDCl3): δ = 10.10–10.07 (m, 1 H), 7.39–7.30 (m, 5 H),
5.14–5.12 (m, 2 H), 4.88–4.39 (m, 4 H), 4.06–3.50 (m, 4 H); con-
formers and hydrate formation. 13C NMR (151 MHz, CDCl3): δ =
196.7 (d, J = 33.2 Hz), 196.4 (d, J = 33.0 Hz), 154.6, 154.4, 136.3,
136.0, 128.6, 128.5, 128.5, 128.3, 128.2, 128.1, 128.0, 97.6, 97.5,
96.3, 96.2, 96.1, 96.0, 76.4–76.2 (m), 74.2–73.7 (m), 73.2–72.8
(m), 67.5, 67.3, 55.9, 55.7, 55.1, 54.9, 53.4, 52.9, 51.8–51.0 (m),
50.9, 29.8–29.0 (m); conformers. 19F NMR (471 MHz, CDCl3): δ =
–173.08, –174.16; conformers. ESI-HRMS: m/z calcd for
22
C
15H17FNO4 [MH+]: 294.1136; found: 294.1137. [α]D +9.9 (c
0.27, CHCl3, 79% ee).
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D