Lithiation-induced migrations from nitrogen to carbon in terminal aziridines

Article abstract of DOI:10.1002/anie.200604920

(Chemical Equation Presented) Benefiting from deprotection: Lithium 2,2,6,6-tetramethylpiperidide induces N-Boc or N-phosphonate terminal aziridines to undergo regio- and stereoselective N-to-C migration of the protecting group, giving synthetically valua

Full text of DOI:10.1002/anie.200604920

Angewandte
Chemie
DOI: 10.1002/anie.200604920
Protecting-Group Migration
Lithiation-Induced Migrations from Nitrogen to Carbon in Terminal
Aziridines**
David M. Hodgson,* Philip G. Humphreys, Zhaoqing Xu, and John G. Ward
Aziridines are receiving increasing research interest, and a
variety of methods have been recently developed for the
synthesis of terminal aziridines 1 (PG = protecting group) in
racemic or enantiopure form.[1,2] We have been studying the
chemistry of lithiated N-Bus (Bus = tert-butylsulfonyl) termi-
nal aziridines 2[3] and recently reported their dimerization to
give 2-ene-1,4-diamines,[4] their intramolecular cyclopropana-
tion (where R bears an unsaturated functional group),[5] and
their trapping of an external electrophile to give 2,3-
disubstituted aziridines (e.g. 3).[6] Herein, we communicate
that lithiated terminal aziridines bearing a protecting group
on the nitrogen atom alternative to organosulfonyl (alkoxy-
carbonyl, dialkoxy phosphonate) display a profoundly differ-
Scheme 1. Reaction of terminal aziridines with lithium 2,2,6,6-tetrame-
thylpiperidide (LTMP).
ent reactivity mode of significant synthetic utility.
We earlier established that the Bus protecting group was
the most suitable in our lithiation/electrophile-trapping
reactions. Although in situ lithiation/silylation of the N-Boc
(Boc = tert-butoxycarbonyl) aziridine of 1-hexene was suc-
cessful, when quenching with external deuterium (CD3OD)
was attempted, the expected trans-deuterated N-Boc aziri-
As the above transformation proceeds from readily
available N-Boc terminal aziridines to give the synthetically
useful aziridinylester motif[7] with concomitant N-deprotec-
tion, we decided to investigate the process in more detail.
Although lithiation-induced N-to-C 1,2-shifts are known,[8] to
the best of our knowledge only a single isolated example of
this type of N-Boc aziridine 1,2-migration has been previously
noted.[8f] In this latter work, the N-Boc aziridine of styrene
was treated with sBuLi in THF at À988C to give a phenyl-
stabilized a-lithiated aziridine, which underwent migration to
give 2-phenyl-2-Boc aziridine (90%). Application of these
reported conditions to aziridine 5a gave aziridinylester 4a in
56% yield, together with several unidentified by-products.
With LTMP the yield of 4a could be increased to 90% by
leaving the reaction at À788C for 90 min before quenching
(conditions otherwise as specified in Scheme 1). These latter
conditions were then applied to a variety of terminal
aziridines to assess the scope of the method (Table 1).
À
dine was not observed. Instead, a single diastereomer of N H
aziridinylester 4a was isolated (Scheme 1), along with 50%
recovered starting material with 0% D incorporation. The
stereochemistry of ester 4a was initially assigned trans based
on the previously observed trans-selective lithiation by the
sterically demanding LTMP with N-Boc and N-Bus aziri-
dines,[6] and this assignment was later supported by crystallo-
graphic analysis of a related aziridinylester prepared by
similar chemistry (vide infra).
[*] Prof. D. M. Hodgson, P. G. Humphreys, Dr. Z. Xu
Department of Chemistry
University of Oxford
Complications were not observed from potential allylic
deprotonation,[9] cyclopropanation,[5] or benzylic deprotona-
tion (Table 1, entries 2 and 3). X-ray crystallographic analysis
of 4c supported the assigned trans stereochemistry.[10] Distal
and proximal protected alcohols were tolerated (Table 1,
entries 4 and 5), as was a potentially eliminable primary
chloride (Table 1, entry 6). A 2,2,3-trisubstituted aziridinyles-
ter could also be accessed (Table 1, entry 7), although in this
case warming to 08C was required for reaction to occur.
Importantly for asymmetric synthesis, no degradation of ee
was observed under the reaction conditions (Table 1, entry 8,
determined by chiral HPLC analysis of the 2,4-dinitrobenzoyl
derivative).[10] Attempted reaction of a 2,3-disubstituted
aziridine (N-Boc aziridine of cyclohexene) only led to
return of starting material, presumably owing to unfavorable
steric interactions. When the latter reaction mixture was
allowed to warm from À788C to 08C, a mixture of starting
Chemistry Research Laboratory
Mansfield Road, Oxford, OX1 3TA (UK)
Fax: (+44)1865-285002
http://users.ox.ac.uk/~dmhgroup
E-mail: david.hodgson@chem.ox.ac.uk
Dr. J. G. Ward
Neurology & GI Centre of Excellence for Drug Discovery
GlaxoSmithKline
New Frontiers Science Park
Third Avenue, Harlow, Essex CM19 5AW (UK)
[**] We thank the EPSRC and GlaxoSmithKline for a CASE award (to
P.G.H.), the Royal Society for an International Incoming Fellowship
(to Z.X.), the EPSRC National Mass Spectrometry Service (Swan-
sea) for mass spectra, S. P. Hughes for preliminary observations,
and Prof. W. Clegg and Dr. L. Russo (University of Newcastle) for
GSK-funded X-ray crystallographic analysis.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 2245 –2248
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2245

Communications
Table 1: Aziridinylester synthesis.
cally those reactions required higher (> À788C) temperatures
and migration is the rate-limiting step.[8b] As the current
reaction is complete within 90 min at À788C and the rate-
limiting step is lithiation,[10] the process may proceed from a
lithiated aziridine by intramolecular attack at the carbamoyl
carbonyl.[8f,11]
Entry Aziridine 5
Aziridinylester 4
Yield [%]
4a 90
À
tert-Butyl-protected N H aziridinylesters accessed by this
1
2
3
methodology can undergo a variety of synthetically useful
subsequent transformations (see Scheme 3). Regioselective
4b 86
4c 79
4[a]
5[a]
6
4d 87
4e 90
4 f 84
7
8
4g 67[b]
(+)-4b 84
Scheme 3. Aziridinylester transformations: a) Raney Ni, H2, EtOH, RT,
12 h, 99%; b) N-bromosuccinimide, CH2Cl2, RT, 20 h, 95%; c) DBU,
toluene, 1108C, 10 h, 92%; d) (COCl)2, DMSO, Et3N, CH2Cl2, À788C
to RT, 5 h, 92%. DMSO=dimethyl sulfoxide.
[a] TBS=tert-butyldimethylsilyl. [b] Reaction mixture was allowed to
warm from À788C to 08C upon addition of reagents and stirred for
90 min.
material and decomposed material was obtained, whereas
warming to room temperature resulted in only decomposi-
tion.
Given the propensity of related lithiated terminal azir-
idines to dimerize (likely as a result of aggregated species),[4]
we sought to establish whether the present migration reaction
was inter- or intramolecular by a crossover experiment
(Scheme 2). A 1:1 mixture of aziridines 5 f and 6 was treated
hydrogenolysis[12] of aziridinylester 4b gave protected b-
amino acid 8 in quantitative yield. b-Amino acids are of
interest owing to their stability against proteolytic degrada-
tion in vivo which can make them suitable for incorporation in
drug candidates.[13] Oxidative cycloamination of the tethered
olefin of 4b with N-bromosuccinimide[14] gave azabicycle 9.
Subsequent elimination using 1,8-diazabicyclo[5.4.0]undec-7-
ene (DBU) gave enamine 10 in an excellent overall yield.
Such bicyclic enamines undergo regioselective ring opening to
give substituted pyrrolidines.[14] Regioselective Swern oxida-
tion of aziridinylester 4b occurred in a completely regiose-
lective manner to give azirine 11.[15] Azirines are useful
heterocycles in synthesis, not only as imino dienophiles but
also as conventional electrophiles in the synthesis of more
substituted aziridines.[16]
To further demonstrate the utility of this methodology, a
concise asymmetric synthesis of a stable ester of the unstable
antibiotic natural product azirinomycin (12)[17] was examined
(Scheme 4). Aminolytic kinetic resolution of racemic propyl-
ene oxide using the method of Bartoli and co-workers gave
amino alcohol 13 in excellent yield and enantiopurity
(> 99% ee).[2d,10] Subsequent one-flask tosylation of the
alcohol and ring closure proceeded smoothly to give R-
aziridine 14. Treatment with LTMP under the optimized
migration conditions gave aziridinylester 15, which then
underwent regioselective Swern oxidation to give the tert-
butyl ester of natural (S)-azirinomycin 16.
Aziridinylphosphonates 19 are important sources of
diversely substituted aminophosphonates, and the latter find
significant utility as amino acid surrogates in biologically
active peptides.[1,18] We considered whether the phosphonate
N-to-C migration[19] might be possible with N-phosphonate
Scheme 2. Crossover experiment.
with 3 equivalents of LTMP at À788C for 90 min; the
resulting product mixture was carefully purified and analyzed
for the presence of products arising from crossover of the
protecting groups. The only observed products in the crude
reaction mixture were aziridinylesters 4 f and 7, isolated in
89% and 82% yield, respectively. The lack of crossover
products indicates that the reaction proceeds in an intra-
molecular fashion.
Although cleavage–recombination mechanisms have
been proposed for lithiation-induced N-to-C 1,2-shifts, typi-
2246 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2245 –2248

Angewandte
Chemie
Table 2: Aziridinylphosphonate synthesis.[a]
Entry Aziridine 18
Aziridinylphosphonate 19
Yield
[%]
1
19b 79
2[b]
19c 79
19d 87
19e 95
19 f 87
3
4
5[c]
6
19g 58
7[d]
(À)-19h 89
Scheme 4. Synthesis of (S)-azirinomycin ester 16: a) 2 mol% (R,R)-17,
4 mol% 4-nitrobenzoic acid, tert-butylcarbamate, tBuOMe, RT, 24 h,
99%; b) TsCl, KOH, THF, RT, 24 h, 62%; c) LTMP, THF, À788C,
90 min, 70%; d) (COCl)2, DMSO, Et3N, CH2Cl2, À788C to RT, 12 h,
72%. Ts=toluene-4-sulfonyl.
[a] Reaction time: 2 h at À788C unless indicated otherwise. [b] Cy=
cyclohexyl. [c] Reaction time: 1 h. [d] Reaction time: 4 h.
aziridines to give stereocontrolled access to aziridinylphosph-
onates, in an analogous manner to that reported above.[20]
Application of our earlier conditions for N-Boc migration to
N-phosphonate aziridine 18 (R = C4H9,[21] Scheme 5) gave the
Scheme 6. b-Aminophosphonate synthesis.
the synthesis of a b-amino ester and phosphonate, and a
synthesis of (S)-azirinomycin tert-butyl ester in four steps
from racemic propylene oxide.
Scheme 5. Phosphonate N-to-C migration.
Experimental Section
Representative procedure for migration of N-Boc aziridine: nBuLi
(1.6m in hexanes, 0.94 mL, 1.50 mmol) was added dropwise to a
stirred solution of 2,2,6,6-tetramethylpiperidine (0.25 mL, 1.50 mmol)
in THF (3.8 mL) at À788C under argon. This mixture was allowed to
warm to room temperature for 30 min, then the resulting solution was
recooled to À788C and a solution of aziridine (0.50 mmol) in THF
(1.5 mL) was added dropwise over 1 min. The reaction mixture was
stirred for 90 min at À788C, then saturated aqueous NH4Cl (2 mL)
was added and the flask was warmed to room temperature. The
aqueous phase was extracted with Et2O (3 ” 10 mL). The combined
organic phases were dried (MgSO4) and then concentrated under
reduced pressure. Purification of the residue by column chromatog-
raphy (petroleum ether/Et2O, SiO2) gave the aziridinylester.
desired aziridinylphosphonate 19a in 52% yield, and solely as
the trans diastereomer,[10] along with 42% recovered starting
material. It was subsequently found that increasing the
number of equivalents of LTMP to 5, a more dilute reaction
mixture, and an extended reaction time gave aziridinyl-
phosphonate 19a in 91% yield (Scheme 5).
These latter conditions were then applied to a variety of
N-phosphonate aziridines (Table 2), and a similar selectivity
profile to that seen earlier in the N-Boc series was observed.
b-Aminophosphonates and the corresponding phosphonic
acids have attracted considerable interest as replacements for
natural amino acids in various targets of medicinal inter-
est.[13,18] Hydrogenolytive ring opening of aziridinylester (À)-
19h was accomplished under transfer-hydrogenation condi-
tions[22] to give b-aminophosphonate (+)-20 in 68% yield and
over 99% ee[10] (Scheme 6).
Received: December 5, 2006
Published online: February 15, 2007
Keywords: amino acids · aziridines · lithiation · rearrangement ·
.
synthetic methods
In summary, [1,2]anionic rearrangement in N-Boc and N-
phosphonate terminal aziridines provides a stereocontrolled
access to synthetically valuable aziridinylesters and aziridi-
nylphosphonates under experimentally straightforward con-
ditions. The utility of this chemistry has been demonstrated in
[1] Aziridines and Epoxides in Organic Synthesis (Ed.: A. K.
Yudin), Wiley-VCH, Weinheim, 2006.
[2]a) P. Wessig, J. Schwarz, Synlett 1997, 893 – 894; b) J. U. Jeong, B.
Tao, I. Sagasser, H. Henniges, K. B. Sharpless, J. Am. Chem. Soc.
Angew. Chem. Int. Ed. 2007, 46, 2245 –2248
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2247

Communications
1998, 120, 6844 – 6845; c) S. K. Kim, E. N. Jacobsen, Angew.
[12]F. A. Davis, J. Deng, Y. Zhang, R. C. Haltiwanger, Tetrahedron
2002, 58, 7135 – 7143.
[13] Enantioselective Synthesis of b-Amino Acids, 2nd ed. (Eds.: E.
Juaristi, V. A. Soloshonok), Wiley, New York, 2005.
[14]G. Chen, M. Sasaki, X. Li, A. K. Yudin, J. Org. Chem. 2006, 71,
6067 – 6073.
Chem. 2004, 116, 4042 – 4044; Angew. Chem. Int. Ed. 2004, 43,
3952 – 3954; d) G. Bartoli, M. Bosco, A. Carlone, M. Locatelli, P.
Melchiorre, L. Sambri, Org. Lett. 2004, 6, 3973 – 3975.
[3]a) D. M. Hodgson, C. D. Bray, P. G. Humphreys, Synlett 2006, 1 –
ˇ
22; b) D. M. Hodgson, B. Stefane, T. J. Miles, J. Witherington, J.
Org. Chem. 2006, 71, 8510 – 8515.
[4]D. M. Hodgson, S. M. Miles, Angew. Chem. 2006, 118, 949 – 952;
Angew. Chem. Int. Ed. 2006, 45, 935 – 938.
[5]D. M. Hodgson, P. G. Humphreys, J. G. Ward, Org. Lett. 2006, 8,
995 – 998.
[6]D. M. Hodgson, P. G. Humphreys, J. G. Ward, Org. Lett. 2005, 7,
1153 – 1156.
[15]L. Gentilucci, Y. Grijzen, L. Thijs, B. Zwanenburg, Tetrahedron
Lett. 1995, 36, 4665 – 4668.
[16]F. Palacios, A. M. Ochoa de Retana, E. Martinez de Marigorta,
J. M. de Los Santos, Eur. J. Org. Chem. 2001, 2401 – 2414.
[17]a) E. O. Stapley, D. Hendlin, M. Jackson, A. K. Miller, S.
Hernandez, J. M. Mata, J. Antibiot. 1971, 24, 42 – 47; b) T. W.
Miller, E. W. Tristram, F. J. Wolf, J. Antibiot. 1971, 24, 48 – 50.
[18]a) K. Moonen, I. Laureyn, C. V. Stevens, Chem. Rev. 2004, 104,
6177 – 6216; b) F. Palacios, C. Alonso, J. M. de Los Santos, Chem.
Rev. 2005, 105, 899 – 931.
[19]F. Hammerschmidt, M. Hanbauer, J. Org. Chem. 2000, 65, 6121 –
6131.
[20]One example of [1,2]anionic rearrangement of a N-diphenyl-
phosphinoyl aziridine has been observed in an unwanted side
process (25% yield): R. Luisi, V. Capriati, S. Florio, P. D. Cunto,
B. Musio, Tetrahedron 2005, 61, 3251 – 3260.
[7]B. Zwanenburg, P. ten Holte, Top. Curr. Chem. 2001, 216, 93 –
124.
[8]a) E. Vedejs, W. O. Moss, J. Am. Chem. Soc. 1993, 115, 1607 –
1608; b) C. Vogel, Synthesis 1997, 497 – 505; c) N. Kise, H. Ozaki,
H. Terui, K. Ohya, N. Ueda, Tetrahedron Lett. 2001, 42, 7637 –
7639; d) K. W. Kells, A. Ncube, J. M. Chong, Tetrahedron 2004,
60, 2247 – 2257; e) N. Dieltiens, C. V. Stevens, K. G. R. Massche-
lein, T. Rammeloo, Tetrahedron 2005, 61, 6749 – 6756; f) V.
Capriati, S. Florio, R. Luisi, B. Musio, Org. Lett. 2005, 7, 3749 –
3752.
[9]A. Mordini, D. Peruzzi, F. Russo, M. Valacchi, G. Reginato, A.
Brandi, Tetrahedron 2005, 61, 3349 – 3360.
[10]See the Supporting Information for details.
[11]J. J. Eisch, S. K. Dua, C. A. Kovacs, J. Org. Chem. 1987, 52, 4437 –
4444.
[21]K. Osowska-Pacewicka, A. Zwierzak, J. Prakt. Chem. 1986, 328,
441 – 444.
[22]F. Palacios, D. Aparicio, A. M. Ochoa de Retana, J. M. de Los
Santos, J. I. Gil, R. L. de Munain, Tetrahedron: Asymmetry 2003,
14, 689 – 700.
2248 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2245 –2248