One-pot synthesis of chiral azetidines from chloroaldehyde and chiral amines

Article abstract of DOI:10.1016/j.tetlet.2011.05.027

A variety of chiral azetidinepiperidines have been synthesized utilizing an expedient one-pot union of piperidine chloroaldehyde with chiral amines. This two step one-pot procedure provides access to an interesting set of compounds that retain the chiral

Full text of DOI:10.1016/j.tetlet.2011.05.027

Tetrahedron Letters 52 (2011) 3618–3620
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
One-pot synthesis of chiral azetidines from chloroaldehyde and chiral amines
⇑
Suvi T. M. Orr (nee Simila) , Shawn Cabral, Dilinie P. Fernando, Teresa Makowski
Pfizer Global Research and Development, Eastern Point Road, Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A variety of chiral azetidinepiperidines have been synthesized utilizing an expedient one-pot union of
piperidine chloroaldehyde with chiral amines. This two step one-pot procedure provides access to an
interesting set of compounds that retain the chiral purity of the starting chiral amine.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 2 April 2011
Revised 21 April 2011
Accepted 5 May 2011
Available online 14 May 2011
Keywords:
Azetidines
Chiral amines
Reductive amination
Constrained cyclic amines are common structural fragments in
drug molecules, such as in Chantix™ (varenicline).1 These types of
systems provide rigid 3-dimensional structures in low molecular
weight frameworks. The spirocyclic azetidine framework 3 pro-
vides an example of such a system that has not been utilized in
drug discovery until recently,2,3 which is likely due at least in part
to the synthetic inaccessibility.4–6 Azetidine rings can be prepared
by a number of general routes including reduction of b-lactams,7
nucleophilic substitutions with amines,8 cycloadditions,9 and rear-
rangements.4 We herein report a convergent and scalable synthesis
to access structurally challenging chiral spiro-azetidinepiperidines.
The initial route to scaffold 3 involved reductive amination of 1
and 2 in the presence of Ti(iPrO)4 and Na(OAc)3BH providing race-
mic 3 in low yield (30–40%) and purity (Scheme 1). After chiral pre-
parative chromatography, only 15–20% of the desired enantiomer
was obtained which was insufficient to scale up the synthesis to
multigram quantities. An alternative route to 3 involved enantiose-
lective reduction of the in situ generated iminium/enamine inter-
mediate of 1 and 2.10 A wide variety of chiral ruthenium and
rhodium catalysts were screened with the best results achieved
using Rh(COD)2SO3CF3 catalyst and Josiphos SLJ001-1 ligand. This
provided 3 in moderate yield (ꢀ50%) and 84% ee. Again, this meth-
od was deemed not suitable for multigram scale up.
To achieve a scalable process to access templates like 3a, addi-
tional routes were investigated. SN2 displacement of mesylate 4a
or triflate 4b with 1 typically resulted in elimination of the mesy-
late or formation of the unreactive racemic chloride 4c as detected
by chiral HPLC (Scheme 2, Route a). None of the desired 3a was
observed.
A reverse approach using mesylate 5a or triflate 5b with 6 was
also attempted but without success (Scheme 2, Route b). Presum-
ably the neopentyl nature of the electrophilic center is slowing
the process, despite a potential Thorpe–Ingold effect from the qua-
ternary carbon center.
Faced with these disappointments, we were intrigued by the
report from the De Kimpe group where azetidines were accessed
from b-chloroimines (or analogous mesylates).11 We envisioned
that a reaction of chloroaldehyde 9 with the chiral amine 6a in
the presence of a reductant would give the azetidine 3a in one
pot via an iminium intermediate with retention of stereochemis-
try at the starting amine. Chloroaldehyde 9 was prepared as
shown in Scheme 3.6 Deprotonation of commercially available 7
with LDA followed by quenching of the resulting enolate with
chloroiodomethane provided 8 in 66% yield. Reduction of 8 with
LiAlH4 and subsequent Swern oxidation gave 9 in 94% yield over
two steps.
Br
H
*
N
N
Ti(iPrO)4
Br
+
Na(OAc)3BH
30-40%
O
N
Boc
N
Boc
3a
1
2
Scheme 1.
⇑
Corresponding author. Tel.: +1 860 686 4192; fax: +1 860 715 4706.
E-mail address: suvi.orr@pfizer.com (S.T.M. Orr (nee Simila)).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.05.027

S. T. M. Orr (nee Simila) et al. / Tetrahedron Letters 52 (2011) 3618–3620
3619
sired product 3a and the intermediate imine were observed in the
reaction mixture after 3–4 h by LC–MS analysis but upon stirring
for 18 h, complete conversion to 3a was observed. Fully cyclized
H
N
Br
Br
a
+
+
N
Boc
1
R
Table 1
Representative examples
4a
: R = OMs
N
see text
for cond
4b: R = OTf
R1
R2
4c: R = rac-Cl
Cl
*
OR OR
Br
N
1) AcOH, MeOH
2) NaCNBH3
CHO
N
Boc
3a
b
NH2
*
+
R1
R2
80-99%
N
N
Boc
5a:
5b:
N
NH2
Boc
Boc
6a
R = Ms
9
6a-h
3a-h
R = Tf
Entry
Aminea
Product
Yield (%)
Eeb (%)
Scheme 2.
Br
Initial attempts to form 3a involved reaction of 9 with 6a in the
presence of AcOH (2 h, 50 °C) and sodium triacetoxyborohydride at
50 °C resulting in uncyclized reductive amination chloroamine
product 10 in 60% yield. Analysis of crude reaction mixture by
LC–MS indicated presence of a substantial amount of intermediate
tentatively assigned as diacetoxyboron adduct of 6a. Stirring the
mixture under the reductive amination conditions for longer peri-
ods of time or at elevated temperature (70 °C) did not provide any
3a. All attempts to cyclize the chloroamine 10 to the azetidine 3a
using a variety of bases (e.g., KOtBu, LiHMDS, DBU)6 were unsuc-
cessful. In most cases 10 remained unreactive or decomposed upon
heating the reaction mixture.
Success was finally achieved by treating 9 and 6a with AcOH in
methanol at 50 °C for 2 h to preform the imine (as observed by LC–
MS), followed by addition of two equivalents of sodium cyanoboro-
hydride at 70 °C, and stirring at this temperature for 18 h (Scheme
3), in order to drive the reaction to completion.12 A mixtureof the de-
1
3a
99
>99
NH2
6a
2
3
3b
3c
90
65
>99
72
NH2
6b
MeO2C
O
NH2
6c
4
5
3d
3e
87
76
98
NH2
6d
Cl
CO2Me
N
Boc
7
LiAlH4;
Swern
CO2Me
LDA
Cl
>99
94%
(2 steps)
NH2
6e
Boc
I
Cl
N
Boc
66%
Br
8
N
Cl
CHO
6
3f
73
>99
Br
one-pot
NH2
6f
N
N
Boc
3a
+
N
H2N
Boc
Boc
N
9
6a
7
8
3g
3h
70
98
>99
50
NH2
6g
Br
NH
Cl
O
N
H
NH2
6h
N
Boc
10
a
The starting amine ees were +98% based on either information on commercial
materials (6b–d, 6f–h) or +96% on Pfizer internal compounds 6a and 6e.
b
The ees were determined by chiral HPLC by comparison of the chiral sample to
the corresponding racemic compounds.
Scheme 3.

3620
S. T. M. Orr (nee Simila) et al. / Tetrahedron Letters 52 (2011) 3618–3620
2. Smith, E. M.; Sorota, S.; Kim, H. M.; McKittrick, B. A.; Nechuta, T. L.; Bennett, C.;
Knutson, C.; Burnett, D. A.; Kieselgof, J.; Tan, Z.; Ringden, D.; Bridal, T.; Zhou, X.;
Jia, Y.-P.; Dong, Z.; Mullins, D.; Zhang, X.; Priestley, T.; Correll, C. C.; Tulshian,
D.; Czarniecki, M.; Greenlee, W. J. Bioorg. Med. Chem. Lett. 2010, 20, 4602–4606.
3. Xiao, D.; Palani, A.; Aslanian, R.; McKittrick, B. A.; McPhail, A. T.; Correll, C. C.;
Phelps, P. T.; Anthes, J. C.; Ringden, D. Bioorg. Med. Chem. Lett. 2009, 19, 783–
787.
4. For a recent review on azetidine synthesis, see: Brandi, A.; Cicchi, S.; Cordero, F.
M. Chem. Rev. 2008, 108, 3988–4035. and references cited therein.
5. Hillier, M. C.; Chen, C. J. Org. Chem. 2006, 71, 7885–7887.
product 3a was isolated in near quantitative yield and in identical
enantiomeric excess (ee) (96–97%) to the starting amine 6a. Under
these conditions we presume that ring closure to form an azetidini-
um ion intermediate and subsequent reduction proceeds more
quickly than direct reduction of the acyclic iminium ion that is, ob-
served with sodium triacetoxyborohydride. Formation ofa cyanobo-
ron adduct with the amine 6a was detected in minor amounts by
LCMS, but not to the extent seen before with sodium triacetoxyboro-
hydride. The ee of 3a could be further enriched (up to 99% ee) by
recrystallization of 3a from hot acetonitrile or isopropanol.
This reaction scheme was expanded to encompass a variety of
chiral amines and the results are shown in Table 1.13 In some
examples, purification of the crude products by crystallization
from hot ethyl acetate/heptanes provided enantiomerically pure
(>99%) final compounds.
The enantiomeric excess was retained in examples where the
chiral amine did not have enolizable protons adjacent to the chiral
center (entries 1, 2, 4, 6 and 7). In entries 3 and 8, where an enol-
izable proton was present, some scrambling was observed with the
ee eroded to 50–72%.
In conclusion, we have developed a concise route to enantiome-
rically pure spiro-azetidinepiperidines which involves treatment of
a piperidine chloroaldehyde with a chiral amine in the presence of
AcOH and NaCNBH3. The enantiopurity of the chiral amine was
retained in the course of the reaction and the azetidine–piperidine
product could be further enantio-enriched by recrystallization.
6. Hamza, D.; Stocks, M. J.; Décor, A.; Pairaudeau, G.; Stonehouse, J. P. Synlett
2007, 2584–2586.
7. For examples, see: (a) Ojima, I.; Zhao, M.; Yamato, T.; Nakahashi, K.; Yamashita,
M.; Abe, R. J. Org. Chem. 1991, 56, 5263–5277; (b) Ojima, I.; Nakashi, K.;
Brandstadter, S. M.; Hatanaka, N. J. Am. Chem. Soc. 1987, 109, 1798–1805.
8. For examples, see: (a) Pennings, M. L. M.; Reinhoudt, D. N. J. Org. Chem. 1983,
48, 4043–4048; (b) van Elburg, P. A.; Reinhoudt, D. N.; Harkema, S.; van
Hummel, G. J. Tetrahedron Lett. 1985, 26, 2809–2812; (c) Ju, Y.; Varma, R. S. J.
Org. Chem. 2006, 71, 135–141.
9. For examples, see: (a) Schrader, T.; Steglich, W. Synthesis 1990, 12, 1153–1156;
(b) Fischer, G.; Fritz, H.; Rihs, G.; Hunkler, D.; Exner, K.; Knothe, L.; Prinzbach, H.
Eur. J. Org. Chem. 2000, 5, 743–762; (c) Dave, P. R.; Duddu, R.; Li, J.; Surapaneni,
R.; Gilardi, R. Tetrahedron Lett. 1998, 39, 5481–5484.
10. For a recent review of enantioselective enamide and imine reductions, see:
Nugent, T. C.; El-Shazly, M. Adv. Synth. Catal. 2010, 352, 753–819.
11. (a) De Kimpe, N.; Boeykens, M.; Tourwé, D. Tetrahedron 1998, 54, 2619–2630;
(b) Sulmon, P.; De Kimpe, N.; Schamp, N. Tetrahedron 1988, 44, 3653–3670.
12. General procedure for the azetidine–piperidines 3: The chloroaldehyde
(0.5 mmol) was dissolved in anhydrous methanol (3 mL) and the chiral
amine (0.5 mmol) was added, followed by acetic acid (0.5 mmol). The
9
6
mixture was stirred at 50 °C for 2 h whereupon it was cooled to room
temperature and sodium cyanoborohydride (1.0 mmol) was added in portions
and the mixture was stirred at 69 °C for 18 h or until LC–MS analysis showed
complete reaction. The solvent was removed under reduced pressure and the
residue was partitioned between 1 N NaOH (10 mL) and dichloromethane
(10 mL). The layers were separated and the aqueous layer was extracted with
dichloromethane (2 Â 5 mL). The combined organic layers were washed with
brine (10 mL), dried (Na2SO4) and concentrated under reduced pressure to give
a clear oil or a white solid. The crude material was purified by Biotage system
on a 10 g silica gel cartridge by eluting with ethyl acetate in heptanes or by
recrystallization from warm isopropyl alcohol.
Acknowledgments
We express our gratitude to Michael Pamment and Ryan Jones
for scaling up critical intermediates. We are also grateful to Daniel
Virtue for the ee determinations. Additionally, we thank Kimberly
O. Cameron, David Hepworth, Kim McClure, Jeffrey Kohrt and
David Price for helpful discussions.
13. Spectral data for 3b: 1H NMR (CD3OD) d 7.32–7.26 (compd, 5H), 3.32–3.25
(compd, 5H), 3.01–2.95 (m, 2H), 2.90–2.87 (m, 2H), 1.67–1.63 (m, 4H), 1.42 (s,
9H), 1.17 (d, J = 8.0 Hz, 3H); 13C NMR (CD3OD) d 155.0, 143.6, 128.5, 127.4,
127.3, 79.5, 68.9, 63.1, 36.2, 33.3, 28.6, 21.4; LC–MS (M+H) 331.4 calcd for
C
20H30N2O2 + H 331.2. Spectral data for 3f: 1H NMR (CD3OD) d 3.38–3.22
(compd, 7H), 3.10–3.07 (m, 1H), 3.00–2.91 (compd, 4H), 2.90–2.85 (m, 1H),
1.79–1.68 (m, 1H), 1.67–1.59 (compd, 5H), 1.42 (s, 9H), 1.38 (s, 9H); 13C NMR
(CD3OD) d 155.1 155.0, 79.6, 79.3, 67.2, 66.5, 62.6, 49.5, 49.1, 44.7, 36.1, 33.2,
28.7, 28.6; LC–MS (M+H) 396.1 calc for C21H37N3O4 + H 396.3.
References and notes
1. Coe, J. W.; Brooks, P. R.; Vetelino, M. G.; Wirtz, M. C.; Arnold, E. P.; Huang, J.;
Sands, S. B.; Davis, T. J.; Lebel, L. A.; Fox, C. B.; Shrikhande, A.; Heym, J. H.;
Schaeffer, E.; Rollema, H.; Lu, Y.; Mansbach, R. S.; Chambers, C. K.; Rovetti, C. C.;
Schulz, D. W.; Tingley, D., III; O’Neill, B. T. J. Med. Chem. 2005, 48, 3474–3477.