Chirality transfer in azetidinium ylides: An enantioselective route to α-quaternary azetidines

Article abstract of DOI:10.1055/s-0028-1087939

Enantiomerically pure N-allyl azetidinium ions undergo a stereoselective [2,3]-sigmatropic shift to give azetidines with an a-quaternary center. These compounds are direct precursors of 2-alkyl-2-carboxy-azetidines, a new class of constrained α-amino acids. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0028-1087939

LETTER
767
Chirality Transfer in Azetidinium Ylides: An Enantioselective Route
to a-Quaternary Azetidines
C
hiralityTransfe
r
inAzet
u
idinium
Y
lid
n
es o Drouillat, François Couty,* Jérome Marrot
UniverSud Paris, Institut Lavoisier de Versailles, UMR CNRS 8180, Université de Versailles Saint Quentin en Yvelines,
45 Avenue des Etats-Unis, 78035 Versailles Cedex, France
Fax +33(1)39254452; E-mail: couty@chimie.uvsq.fr
Received 2 October 2008
fer was in operation.11 The introduction of an allyl group
is also particularly interesting since it can give rise to a
number of derivatives through cross-metathesis of the
double bond.12 Apart from the stereoisomeric consider-
ations, we were also intrigued to see if the strain of the
four-membered ring would unveil a carbenoid-like reac-
tivity of the azetidinium ylide to give 4, that would logi-
cally evolve to piperidine 5 through intramolecular
cyclopropanation of the double bond.13
Abstract: Enantiomerically pure N-allyl azetidinium ions undergo
a stereoselective [2,3]-sigmatropic shift to give azetidines with an
a-quaternary center. These compounds are direct precursors of
2-alkyl-2-carboxy-azetidines, a new class of constrained a-amino
acids.
Key words: azetidines, nitrogen ylides, a-quaternary amino acids
The synthesis of enantiopure a-substituted prolines is a
field of great interest, due to the exceptional role of pro-
line in the control of the secondary structures of peptides.
Asymmetric syntheses of such amino acids can be
achieved through well-established procedures,1 and
among these, the method based on the concept of self-
reproduction of chirality developed by Seebach et al. is
probably the most popular.2 In contrast, the asymmetric
synthesis of a-quaternary 2-carboxyazetidine (i.e., the
lower homologue of proline) is almost uncovered. Be-
sides, in these cyclic amino acids, the geometrical con-
straint brought by the four-membered ring is of special
interest since it was recently demonstrated that these
amino acids are efficient scaffolds for the induction of g-
turns in short peptides.3 To the best of our knowledge, the
asymmetric synthesis of such amino acids has been
achieved only quite recently through two different meth-
odologies. The first one relies on the chemoselective
reduction4 of a suitably functionalized homochiral azeti-
din-2-one, itself prepared through chiral-auxiliary-medi-
ated asymmetric synthesis,5 and the second one is based
on the elegant concept of memory of chirality.6 This re-
stricted availability of synthetic methodologies reflects
the difficulties encountered in the asymmetric synthesis of
azetidines,7 which also hampers the growing interest in 2-
carboxyazetidine derivatives.8
R"
R"
R"
[2,3]
EWG
EWG
R'
R'
EWG
R'
N
N
N
R
R
R
3
2
1
?
GWE
"R
R"
EWG
R'
••
N
R'
N
R
R
4
5
Scheme 1 Is a stereoselective [2,3]-sigmatropic shift of N-allyl aze-
tidinium ions a viable route to azetidines with an a-quaternary center?
1. allyl bromide, DBU,
Ph
Ph
OH
OH
N
toluene, r.t., 2 h, 58%
2. ClCH2CN, K2CO3, MeCN
reflux, 12 h, 79%
NH2
CN
6
7
SOCl2, CH2Cl2
reflux, 2 h
89%
Ph
N
KOt-Bu (2 equiv)
THF, r.t., 3 h
Ph
Cl
CN
9 (>95% de)
MeOTf, CH2Cl2, r.t., 2 h
79%
N
CN
In continuation of our interest in the chemistry of azetidin-
ic amino acids,9 and in the reactivity of azetidinium
ylides,10 we decided to investigate whether enantiopure a-
quaternary azetidines 3 would be accessible through a
[2,3]-sigmatropic shift involving azetidinium ylide 2
(Scheme 1). Such a transformation has been studied by
West with homologous proline derivatives, and he was
able to demonstrate that an efficient N→C chirality trans-
8
94%
Ph
N
Ph
N+
Ph
N+
allyl triflate
69%
CN
CN
TfO–
CN
TfO-–
11 (>95% de)
12
10 (>95% de)
SYNLETT 2009, No. 5, pp 0767–0770
Advanced online publication: 24.02.2009
x
x.
x
x.
2
0
0
9
Scheme 2 Synthesis of epimeric azetidinium ions 10 and 11
DOI: 10.1055/s-0028-1087939; Art ID: D34608ST
© Georg Thieme Verlag Stuttgart · New York

768
B. Drouillat et al.
LETTER
In order to address the question of reactivity, and for a hand. This is indicative of a rapid inversion of the corre-
convenient examination of the stereoselectivity, we first sponding anion, prior to rearrangement. Secondly, al-
synthesized N-allyl azetidinium ions 10 and 11, epimeric though the configuration of the nitrogen atom is indeed
at the nitrogen stereocenter (Scheme 2). The former com- the main parameter which controls diastereoselectivity,
pound was prepared from (1R,2S)-norephedrine (6) which this latter can be influenced by the substitution of the alk-
was first N-allylated and next reacted with chloroaceto- ene, since 19 was obtained as a mixture of epimers. In this
nitrile. The resulting amino alcohol was chlorinated and particular case, we assume that this mixture corresponds
cyclized into 9. It should be noted, as shown by TLC, that to epimers at C-2 since they are produced exclusively
the stereoselectivity of this reaction results from thermo- from a [2,3]-sigmatropic shift and not a Stevens rear-
dynamic control, favoring the depicted isomer 9, which rangement.21 As a matter of fact, this [2,3]-sigmatropic re-
was isolated in 79% yield. Methylation of this compound arrangement is expected to control the configuration of
with methyltrifluoromethylsulfonate then gave azetidini- the benzylic carbon, due to the defined stereochemistry of
um ion 10 with high stereoselectivity, similarly to related the starting alkene. This possibility could not, however, be
substrates.14 On the other hand, N-allylation of known15 ruled out unambiguously by NOE experiments. The unex-
azetidine 12 with a freshly prepared solution of pected absence of the reaction in the case of substrate 23
allyltrifluoromethylsulfonate16 gave epimeric azetidinium also reflects the sensitivity of this reaction towards steric
ion 11 with high diastereoselectivity.
crowding. In this case, either the ylide is not produced, or
it does not react and gives back 23 by subsequent protona-
tion. Entries 5 and 6 demonstrate that other anion-stabiliz-
ing moieties, such as an ester, can also be used
successfully in this reaction, but the azetidinium 25 was
obtained only in low yield, accompanied by undesired b-
elimination product, a side reaction which can only occur
with 2-alkyl-substituted azetidines.22 In the case of com-
pound 29, the absolute configuration of the newly created
stereocenter was confirmed by X-ray crystallography of
the derived azetidinium salt, whose structure23 is depicted
in Figure 1. Finally, entry 7 shows that an attempt to trans-
fer a benzyl group through a Stevens rearrangement led
only to an intractable mixture of products.
In order to test the viability of the sigmatropic shift, these
epimers were reacted with KOt-Bu in THF and afforded
the expected azetidines 13 and 14 with high yields and se-
lectivities. The absolute configuration of the newly creat-
ed quaternary stereocenter in 14 was determined by NOE
experiments, as depicted in Scheme 3, and this configura-
tion agrees with the work of West et al. in the homologous
series, who observed a delivery of the allyl group to the
same face, in a concerted manner.11 We were not able to
isolate under various experimental conditions [use of oth-
er bases, addition of Cu(II) or Rh(II) catalyst] any cyclo-
propyl piperidine 5 that would be diagnostic of the
carbenoid character of the intermediate ylide.
Ph
KOt-Bu, THF,
–78 °C to r.t., 2 h
CN
10
93%
N
13 (96% de)
Ph
NOE
H
KOt-Bu, THF,
–78 °C to r.t., 2 h
11
73%
N
CN
14 (80% de)
Scheme 3 An efficient N→C chirality transfer in azetidinium ylides
Encouraged by these results, and in order to examine the
scope of this rearrangement, a small series of stereo-
defined azetidines was synthesized and transformed into
azetidinium salts by quaternarization with either methyl-
or allyltrifluoromethanesulfonate. These salts were next
reacted with KOt-Bu following the conditions depicted in
Scheme 3. Table 1 summarizes the outcome of these reac-
tions.
Figure 1 The ORTEP structure of azetidinium triflate derived from 29
The reactivity of the produced azetidines with a quaterna-
ry stereocenter was also briefly examined. First, we were
intrigued to know whether the strained amino nitrile moi-
ety in 13 could be reduced.24 As a matter of fact, com-
pound 12, in which a secondary amino nitrile is present,
does not react at all with NaBH4 to give the corresponding
reduced product, and this was attributed to the difficult
generation of an iminium ion in 2-cyano azetidines. On
the other hand, compound 13 reacted sluggishly with
NaBH4 under the same conditions to give a 1:1 mixture of
33 and 14, the epimerized starting compound at the amino
Entries 1–3 in Table 1 illustrate some important points.
First, the relative configuration of the amino nitrile moiety
in the starting azetidinium ions does not influence at all
the stereochemical outcome of the rearrangement, since
exactly the same distribution of epimers is obtained start-
ing from 11 and 16 on one hand and 18 and 21 on the other
Synlett 2009, No. 5, 767–770 © Thieme Stuttgart · New York

LETTER
Chirality Transfer in Azetidinium Ylides
769
Table 1 Scope of the [2,3]-Sigmatropic Rearrangement of N-Allyl Azetidinium Trifluoromethanesulfonates
Entry
1
Starting azetidine
Azetidinium salts
Product(s)
Yield (%)a
Ph
Ph
Ph
CN
TfO–
CN
N
74
N+
N
CN
1515
14
16 (69%)
Ph
Ph
Ph
CN
N
CN
Ph
CN
2
3
50b
N+
TfO–
N
Ph
Ph
19
18 (89%)
1717
Ph
Ph
Ph
CN
N
CN
Ph
71b
CN
TfO–
N+
N
Ph
Ph
19
21 (78%)
2017
Ph
Ph
CN
TfO–
N+
CN
N
4
no reaction
–
2217
23 (77%)
Ph
Ph
Ph
COOEt
TfO–
5
6
7
COOEt
N
68
64
–
N+
N
COOEt
2418
25 (37%)c
26
COOEt
TfO–
COOEt
N
N+
COOEt
N
Ph
2719
Ph
Ph
28 (58%)
29
Ph
Ph
CN
N
decomposition
CN
+TfO–
N
Bn
Bn
3020
31 (95%)
a Yield of isolated product.
b This compound was obtained as a 1:1 mixture of epimers (see text).
c Elimination product (52%) was isolated in this experiment (see text).
nitrile stereocenter (Scheme 4). This unexpected epimeri- The epimerization of 13 to 14 reflects the greater stability
zation was confirmed by a third experiment: reaction of of 14 compared to 13. As a matter of fact, in 14, the trans
13 with sodium cyanoborohydride resulted in complete diaxial disposition of the nitrile group and the lone pair of
epimerization of this stereocenter to give 14 in good iso- the nitrogen atom allows a stabilizing anomeric effect to
lated yield. This last transformation can only be explained occur by overlap of the n orbital at the nitrogen atom and
by the formation of an intermediate iminium ion from the the s* orbital of the C–CN bond.
quaternary amino nitrile, which is probably easier to form
here than from 12.
In summary, we have developed an efficient synthesis of
stereodefined a-quaternary azetidines,25 which are direct
precursors of a new class of constrained a-amino acids.
Synlett 2009, No. 5, 767–770 © Thieme Stuttgart · New York

770
B. Drouillat et al.
LETTER
(15) Agami, C.; Couty, F.; Evano, G. Tetrahedron: Asymmetry
2002, 297.
(16) Beard, C. D.; Baum, K.; Grakauskas, V. J. Org. Chem. 1973,
38, 3673.
Ph
N
NaBH4, EtOH,
reflux, 12 h
no reaction
CN
CN
(17) Compounds 17 and 20 were obtained as a 6:4 ratio of
epimers by anionic cyclization of the corresponding
chloride, obtained following a similar synthetic sequence as
the one described in Scheme 1. Compound 22 was obtained
as the major epimer (7:3 ratio) following Scheme 1.
(18) Couty, F.; Durrat, F.; Prim, D. Tetrahedron Lett. 2003, 44,
5209.
(19) This ester was prepared by reaction of the corresponding
nitrile (ref. 9a) in a mixture of EtOH–H2SO4 (yield 83%).
(20) Drouillat, B.; Couty, F.; David, O.; Evano, G.; Marrot, J.
Synlett 2008, 1345.
12
Ph
Ph
N
NaBH4, EtOH,
reflux, 12 h
+
14
64%
N
13 (96% de)
33 (72% de)
NaBH3CN, EtOH,
r.t., 48 h
N
N
Ph
Ph
CN
(21) Couty, F.; Durrat, F.; Evano, G.; Marrot, J. Eur. J. Org.
CN
84%
Chem. 2006, 4214.
(22) Cospito, G.; Illuminati, G.; Lilloci, C.; Petride, H. J. Org.
Chem. 1981, 46, 2944.
(23) The crystal structure data have been deposited at the
Cambridge Crystallographic Data Centre and allocated the
deposition number CCDC 703601.
14 (96% de)
13 (96% de)
Scheme 4 An iminium ion can be generated from a-quaternary amino
nitriles in azetidines
(24) For a recent review on the reductive decyanation, see:
Mattalia, J.-M.; Marci-Delapierre, C.; Hazimeh, H.; Chanon,
M. ARKIVOC 2006, (iv), 90.
(25) General Procedure for Rearrangement of Azetidinium
Triflates
Acknowledgment
The authors thank the CNRS for financial support. Ms K. Gamee
and A. Nigron are acknowledged for technical assistance.
The following procedure for the preparation of azetidine 13
is representative. To a solution of azetidinium triflate 10
(823 mg, 2.19 mmol) in dry THF (40 mL), cooled at –78 °C
was added in one portion KOt-Bu (300 mg, 2.67 mmol). The
reaction mixture was allowed to reach 0 °C over 3 h and was
quenched by addition of H2O and Et2O. The reaction mixture
was extracted with Et2O, the combined organic layers were
washed with brine, dried over MgSO4, and concentrated
under reduced pressure. The crude residue was examined by
1H NMR and showed a diastereomeric ratio of 98:2.
Purification by flash chromatography (cyclohexane–EtOAc,
8:2) gave 13 as a colorless oil (461mg, 93%).
Selected Data
References and Notes
(1) For a recent review on this topic, see: Calaza, M. I.;
Cativiela, C. Eur. J. Org. Chem. 2008, 3427.
(2) Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem., Int.
Ed. Engl. 1996, 35, 2708.
(3) Baeza, J. L.; Gerona-Navarro, G.; Pérez de Vega, M. J.;
García-López, M. T.; Gonzàles-Muñiz, R.; Martín-
Martínez, M. Tetrahedron Lett. 2007, 48, 3689.
(4) Gerona-Navarro, G.; Bonache, M. A.; Alías, M.;
Pérez de Vega, M. J.; García-López, M. T.; Cativiela, C.;
Gonzàles-Muñiz, R. Tetrahedron Lett. 2004, 45, 2193.
(5) Gerona-Navarro, G.; García-López, M. T.; Gonzàles-Muñiz,
R. J. Org. Chem. 2002, 67, 3953.
(6) Kawabata, T.; Matsuda, S.; Kawakami, S.; Monguchi, D.;
Moriyama, K. J. Am. Chem. Soc. 2006, 128, 15394.
(7) Couty, F.; Evano, G.; Prim, D. Mini-Rev. Org. Chem. 2004,
1, 133.
(8) Couty, F.; Evano, G. Org. Prep. Proced. Int. 2006, 38, 427.
(9) (a) Couty, F.; Evano, G.; Vargas-Sanchez, M.; Bouzas, G.
J. Org. Chem. 2005, 70, 9028. (b) Braüner-Osborne, H.;
Bunch, L.; Chopin, N.; Couty, F.; Evano, G.; Jensen, A. A.;
Kusk, M.; Nielsen, B.; Rabasso, N. Org. Biomol. Chem.
2005, 3, 3926. (c) Mangaleshwaran, S.; Couty, F.; Evano,
G.; Srinivas, B.; Sridhar, R.; Rama Rao, K. ARKIVOC 2007,
(x), 71.
(10) (a) Couty, F.; David, O.; Larmanjat, B.; Marrot, J. J. Org.
Chem. 2006, 72, 1058. (b) Alex, A.; Larmanjat, B.; Marrot,
J.; Couty, F.; David, O. Chem. Commun. 2007, 2500.
(11) Glaeske, K. W.; West, F. G. Org. Lett. 1999, 1, 31.
(12) Lumini, M.; Cordero, F. M.; Pisanechi, F.; Brandi, A. Eur. J.
Org. Chem. 2008, 2817.
25
Compound 13: Rf = 0.48 (cyclohexane–EtOAc, 8:2); [a]D
–13.1 (c 0.33, CHCl3). 1H NMR (300 MHz,CDCl3): d = 1.42
(d, J = 6.6 Hz, 3 H, Me), 2.01–2.12 (m, 1 H, CHHCH=CH2),
2.21–2.32 (m, 1 H, CHHCH=CH2), 2.42 (s, 3 H, NMe),
3.65 (d, J = 5.4 Hz, 1 H, H3), 3.80 (q, J = 5.5 Hz, 1 H, H4),
4.85–5.03 (m, 2 H, CHHCH=CH2), 5.35–5.52 (m, 1 H,
CHHCH=CH2), 7.23–7.35 (m, 5 H, Ar). 13C NMR (75 MHz,
CDCl3): d = 17.2 (CH3), 33.9 (NMe), 36.3 (CH2), 53.3 (C3),
61.9 (C4), 64.9 (C2), 119.9 (CN), 121.1 (CH=CH2), 126.9,
128.3, 129.1 (CHAr), 131.0 (CH=CH2), 135.6 (CqAr). MS
(CI, NH3): m/z = 227.1 (100) [MH+], 200.2 (50) [MH+ –
HCN].
Compound 14; yield 73%, colorless oil; Rf = 0.34 (pentane–
EtOAc, 9:1); [a]D25 –64 (c 1, CH2Cl2). 1H NMR (300 MHz,
CDCl3): d = 1.17 (d, J = 5.8 Hz, 3 H, Me), 2.38 (s, 3 H,
NMe), 2.58 (appt. d, J = 5.3 Hz, 2 H, CH2CH=CH2), 3.15
(d, J = 9.1 Hz, 1 H, H3), 3.35 (q, J = 5.8 Hz, 1 H, H4),
5.11–5.22 (m, 2 H, CHHCH=CH2), 5.70–5.84 (m, 1 H,
CHHCH=CH2), 7.12–7.21 (m, 5 H, Ar). 13C NMR (75 MHz,
CDCl3 MHz): d = 19.6 (CH3), 37.9 (NMe), 43.8 (CH2), 53.9
(C3), 63.5 (C4), 71.2 (C2), 117.2 (CN), 120.2 (CH=CH2),
127.8, 128.3, 128.6 (CHAr), 130.9 (CH=CH2), 135.3
(CqAr). MS (CI, NH3): m/z = 227.1 (100) [MH+], 200.2 (35)
[MH+ – HCN].
(13) Khlebnikov, A. F.; Novikov, M. S.; Kostikov, R. R. Russ.
Chem. Rev. 2005, 74, 171.
(14) Couty, F.; Durrat, F.; Evano, G.; Prim, D. Tetrahedron Lett.
2004, 45, 7525.
Synlett 2009, No. 5, 767–770 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.