Synthesis of new 3,3-dimethoxyazetidine-2-carboxylic acid derivatives

Article abstract of DOI:10.1016/j.tetlet.2004.10.125

Cyclization of γ-amino-α-bromocarboxylic esters resulted in an efficient synthesis of new 3,3-dimethoxyazetidine-2-carboxylates, that is, methyl N-t-butyl-3,3-dimethoxyazetidine-2-carboxylic ester and 3,3-dimethoxyazetidine-2-carboxylic acid, or 3-bromo-4

Full text of DOI:10.1016/j.tetlet.2004.10.125

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 525–529
Synthesis of new 3,3-dimethoxyazetidine-2-carboxylic
acid derivatives
Sven Mangelinckx,a Marc Boeykens,a Maarten Vliegen,b
Johan Van der Eyckenb and Norbert De Kimpea,*
aDepartment of Organic Chemistry, Faculty of Agricultural and Applied Biological Sciences, Ghent University,
Coupure Links 653, B-9000 Ghent, Belgium
bDepartment of Organic Chemistry, Faculty of Sciences, Ghent University, Krijgslaan 281 S4, B-9000 Ghent, Belgium
Received 8 July 2004; revised 21 October 2004; accepted 22 October 2004
Abstract—Cyclization of c-amino-a-bromocarboxylic esters resulted in an efficient synthesis of new 3,3-dimethoxyazetidine-2-carb-
oxylates, that is, methyl N-t-butyl-3,3-dimethoxyazetidine-2-carboxylic ester and 3,3-dimethoxyazetidine-2-carboxylic acid, or 3-
bromo-4,4-dimethoxypyrrolidin-2-ones, depending on the substituent at nitrogen. Reduction of the 3,3-dimethoxyazetidine-2-carb-
oxylates gave the corresponding 3,3-dimethoxy-2-(hydroxymethyl)azetidines. These novel cyclic amino acid derivatives, available on
multigram scale, have a suitably protected carbonyl function at the 3-position, which enables further functionalization.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
Compared to azetidin-2-ones (b-lactams), azetidin-3-
ones are much less widespread.1 The latter have been re-
garded as synthons for other natural N-heterocyclic
four-membered rings, for example, 3-ethylideneazeti-
din-2-carboxylic acid (polyoximic acid).2 More specifi-
cally, methyl 1-benzhydryl-3,3-dimethoxyazetidine-2-
carboxylate has proven to be a useful building block
in the synthesis of 4-alkyl-1-benzhydryl-2-(methoxy-
methyl)azetidin-3-ols.3 These compounds are represent-
atives of the class of 2,4-disubstituted azetidin-3-ols,
which are found in nature as the alkaloids penarisidine
A and B and penazetidine A.4 In the present article we
would like to disclose our results on the synthesis of
3,3-dimethoxyazetidine-2-carboxylic esters and 3,3-
dimethoxyazetidine-2-carboxylic acids, which can be re-
garded as useful new building blocks for other azetidine
derivatives.5 This is the first time that such protected
cyclic amino acids are made accessible.
It has been shown that azetidine-2-carboxylic acid and
its derivatives are easily accessible by a cyclocondensa-
tion reaction between primary amines and a,c-dibromo-
esters,6 or, in an intramolecular version, from base
treatment of c-amino-a-halogenated acids7 or a-(N-tos-
yl)amino-c-halogenated carboxylic esters.8 Initially,
this more direct route towards azetidine-2-carboxylic
acid derivatives through cyclocondensation was exam-
ined for the synthesis of 3,3-dimethoxyazetidine-2-carb-
oxylates. The substrate for this reaction, that is, the a,c-
dihalo ester 3, was prepared starting from the commer-
cial methyl 4-chloro acetoacetate 1 (Scheme 1). The lat-
ter reaction included alkoxybromination of 4-chloro-3-
methoxy-2-butenoate 2, prepared from 1 as described
in the literature.9 Despite the presence of an inductively
activated halogen no reaction could be established be-
tween the a-bromo-c-chloro ester 3 and a primary
amine. Even sodium azide, successfully applied in struc-
turally related cases, was found to be unreactive under a
series of reaction conditions.10 In a second approach,
nitrogen was introduced via nucleophilic substitution
of the allylic halogen in 2 prior to alkoxyhalogenation.11
Treatment of 2 with an excess of t-butylamine in aceto-
nitrile under reflux for 1h afforded pure c-aminocroto-
nate 6 in 94% yield. The excess of primary amine was
used to minimize the reaction time, as prolonged heating
Keywords: Azetidin-3-ones; Azetidine-2-carboxylates; Pyrrolidin-2-ones.
*
Corresponding author. Tel.: +32 9 264 59 51; fax: +32 9 264 62 43;
e-mail: norbert.dekimpe@ugent.be
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.125

526
S. Mangelinckx et al. / Tetrahedron Letters 46 (2005) 525–529
N3
Br
1 equiv. i-PrNH2
5 equiv. NaN3
MeO OMe
MeO
MeO
MeO
MeO
X
CO2Me
X
CO2Me
2 equiv. Et3N
DMSO/ 65°C
CO2Me
CH2Cl2/ ∆
N
3d
Cl
Cl
5
3 (91%)
4
1 equiv. NBS
MeOH/ rt/ 30'
1. HCl/MeOH
rt/ 15h
O
5 equiv. t-BuNH2
MeO
MeO
Cl
Cl
CO2Me
CO2Me
CO2Me
2. MeSO3H cat.
100mbar
CH3CN
∆/ 1h
N
H
125-130°C
1
2 (87%)
6 (94%)
5 equiv. t-BuNH2
CH3CN
∆/ 17h
RNH2
MeO
OMe
OMe
H
4 equiv. NaBH4
O
H
N
CO2Me
N
CO2Me
N
R
MeOH
∆/ 2h
9 R = H (ref. 9)
10 R = i-Pr (94%)
7 (60%)
8 (67%)
Scheme 1.
resulted in the formation of the regioisomeric c-imino
butanoate 7 as the sole product, originating from two
double bond migrations.
Br
1. HCl (g)
Et2O
MeO
MeO
MeO
CO2Me
CO2Me
2. 1 equiv. NBS
MeOH
rt/ 1h
N
H
+
-
N
Cl
H
H
The use of the more sterically hindered t-butylamine was
of prime importance in more than one way. In the case
of lower analogues such as isopropylamine and ammo-
nia9 substitution of the allylic halogen was accompanied
by lactam formation. The t-butyl group thus clearly pre-
vented cyclization of the crotonate formed. The same
accounted for similar saturated compounds, like c-amino-
ester 8.12 The latter was the only compound observed
upon reduction of the c-imino butanoate 7 with sodium
borohydride in methanol under reflux.
11
6
2 equiv. t-BuNH2
MeOH
∆/ 2h
MeO OMe
MeO OMe
N
2 equiv. LiAlH4
CO2Me
Et2O/ ∆
2h
N
OH
13 (93%)
12 (74%)
Scheme 2.
The vicinal methoxybromination of the amino crotonate
6 proceeded smoothly, provided the hydrochloride salt
of the free base 6 was applied. Since the t-alkyl substi-
tuent precluded lactam formation, reflux of 11 in metha-
nol in the presence of t-butylamine exclusively afforded
the desired azetidinecarboxylic ester 12 (Scheme 2).11
The electrophilic addition and consecutive ring closure
could be run as a one pot reaction without isolation of
the intermediate c-amino ester 11. Even though this ap-
proach was restricted to an amine bearing a t-butyl sub-
stituent, it was an important achievement. The reactions
involved are feasible on a large scale. Substantial
amounts of the azetidine 12 (38g) in a 60% overall yield
starting from methyl 4-chloro acetoacetate 1 could be
obtained. Reduction of the azetidine ester 12 with lith-
ium aluminium hydride finally afforded 3,3-dimethoxy-
2-(hydroxymethyl)azetidine 13 in 93% yield.
143 generated the iminophosphorane 15,13 which precipi-
tated from pentane as a white amorphous solid (Scheme
3). By hydrolysis of 15 with 1equiv of water in THF, a
primary amino group was introduced13 while circum-
venting lactam formation in contrast to direct displace-
ment with ammonia.9 Somewhat disappointingly, an
aza-Wittig reaction of the phosphorane 15 with aro-
matic aldehydes was always accompanied by isomeriza-
tion of the imino compound to the 2-azadiene 17.
Alkoxyhalogenation of N-(arylidene)aminobutenoate
18 and hydride-induced cyclization would have lead
again to an azetidine-2-carboxylic ester bearing a
removable N-substituent.
Efforts were then concentrated on 4-azido-2-bromo-3,3-
dimethoxybutanoate 19, obtained by methoxybromina-
tion of c-azidobutenoate 14 (Scheme 4). Reduction of
the azido group with tin(II) chloride in dry methanol
containing dry hydrogen chloride provided the hydro-
chloride salt of the c-amino ester 20. The propensity
After successfully having accomplished the synthesis of
azetidinecarboxylic ester 12, our attention was focussed
on the c-azidobutenoate 14, which was used in the syn-
thesis of methyl 1-benzhydryl-3,3-dimethoxyazetidine-2-
carboxylate.3 Staudinger reduction of c-azidobutenoate

S. Mangelinckx et al. / Tetrahedron Letters 46 (2005) 525–529
527
2 equiv. NaN3
MeO
MeO
Cl
CO2Me
CO2Me
NH3
MeO
N
CH3CN
CO2Me
∆/ 4h
N3
H
2
14 (94%)
C6H5
17
1 equiv. (C6H5)3P
C5H12/ rt/ 1d
X
+
1 equiv. C6H5CHO
1 equiv. H2O
THF/ rt/ 17h
MeO
H2N
MeO
MeO
CO2Me
CO2Me
CO2Me
THF/ rt
H
(C6H5)3P
N
N
C6H5
16 (75%)
15 (84%)
18
Scheme 3.
Br
MeO
MeO
Br
1 equiv. NBS
MeOH/ rt/ 1h
MeO
MeO
MeO
N3
CO2Me
CO2Me
O
N
N3
C6H5
19 (99%)
23 (59%)
14
2 equiv. NaCNBH3
1 equiv. AcOH/ MeOH
0°C/ 2h
1. 2.2 equiv. SOCl2
-10°C/ MeOH/ 10 min
2. 1.5 equiv. SnCl2/ MeOH
-10°C --> rt/ 17h
Br
MeO
MeO
Br
Br
1 equiv. C6H5CHO
2 equiv. NaHCO3
MeO
MeO
MeO
CO2Me
C6H5
CO2Me
O
CH2Cl2/ H2O
MeO
2 equiv. Et3N
CH2Cl2
MgSO4/ rt/ 2h
N
H
rt/ 2h
N
NH2.HCl
20 (93%)
21 (67%)
22
Scheme 4.
of N-unsubstituted c-amino esters to cyclize was mani-
fested by the neutralization of the hydrochloride salt
20 in a biphase liquid system of aqueous sodium bicarb-
onate and dichloromethane. Stirring at room tempera-
ture for 1h was sufficient to obtain the c-lactam 21 as
the sole product. Another possibility to establish the
synthesis of an N-arylidene-c-amino ester entailed the
condensation of 20 with benzaldehyde. The latter was
however not so successful as the condensation with benzo-
phenone imine.3
of the c-azido ester 19 was hydrolyzed in alkaline med-
ium prior to generation of the amino function (Scheme
5).12 In the next step a Staudinger reaction was per-
formed to reduce the azido group. Treatment of the c-
azido acid 24 with triphenylphosphine in THF provided
a simple way to control the course of the reaction. The
intermediate iminophosphorane 25 precipitated from
the THF solution as soon as it was formed and could
be collected as a white amorphous solid. Hydrolysis of
the intermediate 25 with a small excess of base and
simultaneous ring closure then offered the free 3,3-dime-
thoxyazetidine-2-carboxylic acid 26.14
As expected, the presence of triethylamine led to the form-
ation of lactam 21 as side reaction. Only the N-benz-
ylidene-c-aminoester 22 together with the unreacted
benzaldehyde was isolated after aqueous workup. Due
to the polarity of the lactam 21 and the alkaline workup,
none of it was extracted into the organic layer. Reduc-
tion of 22 with NaBH4 in methanol at room temperature
or even with NaCNBH3 at 0ꢁC finally showed that the
propensity for lactam formation is equally valid for c-
amino esters with a primary N-substituent.
Removal of any inorganic material by means of ion ex-
change chromatography (Dowex 50 · 8, H+) led to con-
siderable loss of material. This is probably due to
hydrolysis of the acetal moiety and concomitant ring
opening through a retro-Dieckmann reaction.15 The
reaction sequence, leading to N-(methoxycarbonyl-
methyl)glycine 30, slowly occurred upon storage of the
amino acid at room temperature (Scheme 6). It is not
inconceivable that the column material works as a cata-
lyst and accelerates the reaction. Direct N-protection of
amino acid 26 with a benzyloxycarbonyl group however,
was possible under Schotten–Baumann conditions,
which resulted in carbamate 27 in good yield starting
from c-azido acid 24. Reduction of the amino acid 26
with lithium aluminium hydride finally yielded the free
amino alcohol 28, which could be benzyloxycarbonyl-
ated selectively at nitrogen to afford carbamate 29.
Up to now attention was focussed on the nucleophilic
part of the azetidine precursors. In order to prevent
the amino function from attacking the ester carbonyl
with concomitant lactam formation, the electrophilic
part was rendered less attractive towards nucleophilic
attack. The best way to do this was to increase its elec-
tron density, that is, to protect it as its corresponding
carboxylic acid salt. For this purpose, the ester moiety

528
S. Mangelinckx et al. / Tetrahedron Letters 46 (2005) 525–529
1) 1N NaOH
Br
Br
Br
1 equiv. (C6H5)3P
MeOH/ rt/ 3h
MeO
MeO
MeO
MeO
MeO
MeO
CO2Me
CO2H
CO2
2) H3O+
4 equiv. Et3N
THF/ rt/ 17h
P(C6H5)3
N3
N3
N
19
24 (99%)
25
1) 2.1 equiv. NaOH
THF/ H2O 1/1
∆ / 4h
1) 2.1 equiv. NaOH
THF/ H2O 1/1
∆ / 4h
2) 1 equiv. ClCO2Bn
1.1 equiv. Na2CO3
CH2Cl2, H2O 1/1
rt/ 3h
2) H3O+
MeO OMe
MeO OMe
2 equiv. LiAlH4
CO2H
THF/∆/ 17h
N
H
N
H
OH
28 (51%)
26 (41%)
MeO OMe
1 equiv. ClCO2Bn
1 equiv. NaHCO3
H2O/ rt/ 1h
CO2H
N
CO2Bn
27 (73%)
MeO OMe
N
OH
CO2Bn
29 (64%)
Scheme 5.
4. (a) Kobayashi, J.; Cheng, J.; Ishibashi, M.; Wa¨lchli, M.
R.; Yamamura, S.; Ohizumi, Y. J. Chem. Soc., Perkin
Trans. 1 1991, 1135–1137; (b) Takikawa, H.; Maeda, T.;
Seki, M.; Koshino, H.; Mori, K. J. Chem. Soc., Perkin
Trans. 1 1997, 97–111; (c) Knapp, S.; Dong, Y. Tetrahe-
dron Lett. 1997, 38, 3813–3816.
MeO OMe
MeO2C
N
H
CO2H
CO2H
rt
N
H
26
30
5. For a review on the synthesis of azetidin-3-ols and of other
azetidines, see: (a) Cromwell, N. H.; Phillips, B. Chem.
Rev. 1979, 79, 331–358; (b) De Kimpe, N. Azetidines,
Azetines and Azetes: Monocyclic. In Three and Four-
Membered Rings, with all Fused Systems Containing Four-
Membered Rings; Padwa, A., Katrizky, A. R., Rees,
C. W., Scriven, E. F. V., Eds.; Comprehensive Hetero-
cyclic Chemistry II; Elsevier: Oxford, UK, 1996; Vol. 1B,
Chapter 1.18, pp 507–589.
6. (a) Rodebaugh, R. M.; Cromwell, N. H. J. Heterocycl.
Chem. 1968, 5, 309–311; (b) Rodebaugh, R. M.; Crom-
well, N. H. J. Heterocycl. Chem. 1969, 6, 435–437; (c)
Phillips, B. A.; Cromwell, N. H. J. Heterocycl. Chem.
1973, 10, 795–799; (d) Wasserman, H. H.; Lipshutz, B.;
Tremper, A. W.; Wu, J. S. J. Org. Chem. 1981, 46, 2991–
2999; (e) Kulkarni, S. B.; Rodebaugh, R. M.; Cromwell,
N. H. J. Heterocycl. Chem. 1976, 13, 329–332; (f) Soriano,
D. S.; Podraza, K. F.; Cromwell, N. H. J. Heterocycl.
Chem. 1980, 17, 1389–1392.
7. (a) Fowden, L. Nature (London) 1955, 176, 347–348; (b)
Fowden, L. Biochem. J. 1956, 64, 323–332; (c) Pichat, L.;
Liem, P. N.; Guermont, J. P. Bull. Soc. Chim. Fr. 1968,
4079–4081; (d) Yamada, S.; Emori, T.; Kinoshita, S.;
Okada, H. Agric. Biol. Chem. 1973, 37, 649–652.
8. (a) Chen, T.; Sanjiki, T.; Kato, H.; Ohta, M. Bull. Chem.
Soc. Jpn. 1967, 40, 2398–2401; (b) Miyoshi, M.; Sugano,
H.; Fujii, T.; Ishihara, T.; Yoneda, N. Chem. Lett. 1973,
5–6.
Scheme 6.
The most important advantage of this pathway clearly is
the direct access to the free amino acid 26.
In conclusion, starting from functionalized c-amino
esters, new 3,3-dimethoxyazetidine-2-carboxylates were
made accessible, along with the corresponding 2-
(hydroxymethyl)azetidines obtained by reduction of
the former. These compounds possess the appropriate
azetidine skeleton required for further total synthesis
of several natural products.
Acknowledgements
The authors are indebted to the IWT, the FWO and
Ghent University (GOA) for financial support of this
research.
References and notes
1. Dejaegher, Y.; KuzÕmenok, N. M.; Zvonok, A. M.; De
Kimpe, N. Chem. Rev. 2002, 102, 29–60.
2. (a) Emmer, G. Tetrahedron 1992, 48, 7165–7172; (b)
Hanessian, S.; Fu, J.-M.; Isono, K. Tetrahedron Lett.
1993, 34, 4153–4156; (c) Hanessian, S.; Fu, J.-M.; Chiara,
J.-L.; Di Fabio, R. Tetrahedron Lett. 1993, 34, 4157–4160;
(d) Isono, K.; Asahi, K.; Suzuki, S. J. Am. Chem. Soc.
1969, 91, 7490–7505; (e) Baumann, H.; Duthaler, R. O.
Helv. Chim. Acta 1988, 71, 1035–1041.
9. Duc, L.; McGarrity, J. F.; Meul, T.; Warm, A. Synthesis
1992, 391–394.
10. (a) Micetich, R. G.; Maiti, S. N.; Spevak, P.; Tanaka, M.;
Yamazaki, T.; Ogawa, K. Synthesis 1986, 292–296; (b)
Durst, T.; Koh, K. Tetrahedron Lett. 1992, 33, 6799–6802.
11. Rodebaugh, R. M.; Cromwell, N. H. J. Heterocycl. Chem.
1971, 8, 19–24.
´
12. Khoukhi, N.; Vaultier, M.; Carrie, R. Tetrahedron 1987,
43, 1811–1822.
3. Salgado, A.; Boeykens, M.; Gauthier, C.; Declercq, J.-P.;
De Kimpe, N. Tetrahedron 2002, 58, 2763–2775.

S. Mangelinckx et al. / Tetrahedron Letters 46 (2005) 525–529
529
´
13. Vaultier, M.; Knouzi, N.; Carrie, R. Tetrahedron Lett.
1983, 24, 763–764.
(70eV) m/z (%): no M+, 149 (4), 132 (15), 115 (8), 101 (7),
89 (5), 88 (98), 87 (29), 86 (11), 84 (8), 75 (11), 74 (13), 73
(8), 71 (5), 70 (5), 69 (12), 60 (9), 59 (11), 58 (46), 57 (17),
56 (7), 55 (12), 47 (4), 45 (14), 44 (33), 43 (100), 42 (21), 41
(18), 40 (26).
14. 3,3-Dimethoxyazetidine-2-carboxylic acid 26: 1H NMR
(D2O, 270MHz): d (ppm) 3.20 (s, 3H), 3.32 (s, 3H), 4.04
(d, J = 11.2Hz, 1H), 4.17 (d, J = 11.2Hz, 1H), 4.79 (s,
1H); 13C NMR (D2O, 67.5MHz): d (ppm) 48.8, 49.3, 53.2,
68.7, 97.7, 167.5; IR (KBr, cmÀ1): 3600–2200, 1680; MS
15. Moyer, M. P.; Feldman, P. L.; Rapoport, H. J. Org.
Chem. 1985, 50, 5223–5230.