Dimerization and isomerization reactions of α-lithiated terminal aziridines

Article abstract of DOI:10.1021/jo701901t

(Chemical Equation Presented) The scope of dimerization and isomerization reactions of α-lithiated terminal aziridines is detailed. Regio-and stereoselective deprotonation of simple terminal aziridines with lithium 2,2,6,6-tetramethylpiperidide (LTMP) or lithium dicyclohexylamide (LiNCy ₂) generates trans-α-lithiated terminal aziridines. These latter species can then undergo dimerization or isomerization reactions depending on the nature of the N-protecting group. α-Lithiated terminal aziridines bearing N-alkoxycarbonyl (Boc) protection undergo N- to C-[1,2] migration to give N-H trans-aziridinylesters. In contrast, aziridines bearing N-organosulfonyl [tert-butylsulfonyl (Bus)] protection undergo rapid dimerization to give 2-ene-1,4-diamines or, if a pendant alkene is present, diastereoselective cyclopropanation to give 2-aminobicyclo[3.1.0]hexanes. All of these reactions were used as key steps in the preparation of synthetically and biologically important targets.

Full text of DOI:10.1021/jo701901t

Dimerization and Isomerization Reactions of r-Lithiated Terminal
Aziridines
David M. Hodgson,*,† Philip G. Humphreys,† Steven M. Miles,†
Christopher A. J. Brierley,† and John G. Ward‡
Department of Chemistry, UniVersity of Oxford, Chemistry Research Laboratory, Mansfield Road, Oxford,
OX1 3TA, United Kingdom, and Neurology & GI Centre of Excellence for Drug DiscoVery,
GlaxoSmithKline New Frontiers Science Park, Third AVenue, Harlow, Essex, CM19 5AW, United Kingdom
daVid.hodgson@chem.ox.ac.uk
ReceiVed August 29, 2007
The scope of dimerization and isomerization reactions of R-lithiated terminal aziridines is detailed. Regio-
and stereoselective deprotonation of simple terminal aziridines with lithium 2,2,6,6-tetramethylpiperidide
(LTMP) or lithium dicyclohexylamide (LiNCy2) generates trans-R-lithiated terminal aziridines. These
latter species can then undergo dimerization or isomerization reactions depending on the nature of the
N-protecting group. R-Lithiated terminal aziridines bearing N-alkoxycarbonyl (Boc) protection undergo
N- to C-[1,2] migration to give N-H trans-aziridinylesters. In contrast, aziridines bearing N-organosulfonyl
[tert-butylsulfonyl (Bus)] protection undergo rapid dimerization to give 2-ene-1,4-diamines or, if a pendant
alkene is present, diastereoselective cyclopropanation to give 2-aminobicyclo[3.1.0]hexanes. All of these
reactions were used as key steps in the preparation of synthetically and biologically important targets.
Introduction
fashion to their more thoroughly studied epoxide cousins
(Scheme 1).4
Aziridines are useful synthetic intermediates1 that are seeing
increasing research interest, both in their synthesis and elabora-
tion.2,3 Treatment of a suitably N-protected aziridine 1 (PG )
protecting group) with a hindered lithium amide, or an orga-
nolithium/diamine combination typically results in lithiation of
the aziridine ring to give an R-lithiated aziridine 2, in a similar
SCHEME 1. Aziridine r-Lithiation
† University of Oxford.
In the absence of an anion-stabilizing group lithium carbenoid
2 is rather unstable, which is attributable to the potential for
‡ GlaxoSmithKline.
(1) (a) Deyrup, J. A. In Chemistry of Heterocyclic Compounds; Vol. 42:
Small Ring Heterocycles; Pt. 1: Aziridines, Azirines, Thiiranes, Thiirenes;
Hassner, A., Ed.; Wiley-VCH: Chichester, UK, 1983; pp 1-214. (b)
Sweeney, J. B. Chem. Soc. ReV. 2002, 31, 247-258.
(2) (a) Wessig, P.; Schwarz, J. Synlett 1997, 893-894. (b) Jeong, J. U.;
Tao, B.; Sagasser, I.; Henniges, H.; Sharpless, K. B. J. Am. Chem. Soc.
1998, 120, 6844-6845. (c) Kim, S. K.; Jacobsen, E. N. Angew. Chem.,
Int. Ed. 2004, 43, 3952-3954. (d) Bartoli, G.; Bosco, M.; Carlone, A.;
Locatelli, M.; Melchiorre, P.; Sambri, L. Org. Lett. 2004, 6, 3973-3975.
(3) (a) Aziridines and Epoxides in Organic Synthesis; Yudin, A. K., Ed.;
Wiley-VCH: Weinheim, Germany, 2006. (b) Singh, G. S.; D’hooghe, M.;
De Kimpe, N. Chem. ReV. 2007, 107, 2080-2135.
(4) For recent reviews of lithiated epoxide and aziridine chemistry, see:
(a) Oxiranyl and aziridinyl anions as reactive intermediates in synthetic
organic chemistry. Florio, S., Ed. Tetrahedron 2003, 59, 9683-9864. (b)
Chemla, F.; Vrancken, E. In The Chemistry of Organolithium Reagents;
Rappoport, Z., Marek, I., Eds.; Wiley and Sons: New York, 2004; pp 1165-
1242. (c) Hodgson, D. M.; Bray, C. D. In Aziridines and Epoxides in
Organic Synthesis; Yudin, A. K., Ed.; Wiley-VCH: Weinheim, Germany,
2006; pp 145-184. (d) Hodgson, D. M.; Bray, C. D.; Humphreys, P. G.
Synlett 2006, 1-22. (e) Hodgson, D. M.; Humphreys, P. G.; Hughes, S. P.
Pure. Appl. Chem. 2007, 79, 269-280.
10.1021/jo701901t CCC: $37.00 © 2007 American Chemical Society
Published on Web 11/17/2007
J. Org. Chem. 2007, 72, 10009-10021
10009

Hodgson et al.
R-elimination driven by relief of aziridine ring-strain;5 as such,
most studies to date have focused on aziridines bearing anion-
stabilizing substitution at the site of metallation. The nature of
the substituent alters the reactivity profile of the lithium
carbenoid species markedly: R-lithiated aziridines bearing an
anion-stabilizing group undergo electrophile trapping reactions,4
whereas those derived from disubstituted alkenes have been
shown to react by carbenoid pathways such as C-H insertion,6
reductive elimination,7 and conversion into alkynyl amino
alcohols.7b,8 R-Lithiated aziridines without substitution at the
site of lithiation (i.e., those derived from terminal aziridines)
have received little attention to date. Although there are
examples of R-lithiated aziridines generated by tin-lithium
exchange9 and by activation of the aziridine ring by complex-
ation with a Lewis acid,10 prior to our studies in this area,11
only a single account existed of the direct hydrogen-lithium
exchange reaction to generate an R-lithiated terminal aziridine:
the N-Boc aziridine of propene 3 together with TMEDA and
an excess of Me3SiCl was treated with s-BuLi at -78 °C to
give a trans/cis12 mixture of aziridinylsilanes 4 (Scheme 2).13
terminal epoxides using a mixture of the sterically demanding
lithium 2,2,6,6-tetramethylpiperide (LTMP) and Me3SiCl17 could
be applied to a regio- and diastereoselective synthesis of trans-
R,â-aziridinylsilanes.
The synthesis of terminal N-Boc aziridine 5 was accomplished
by radical mediated amino-bromination of 1-hexene,18 followed
by aziridine ring-closure on treatment with NaH (Scheme 3).
With aziridine 5 in hand, in situ lithiation/silylation was found
to be possible at -78 °C, giving R,â-aziridinylsilane 6 as a
single trans-diastereomer, in 69% yield. In trying to extend this
chemistry further, we investigated the use of external electro-
philes. However, reaction of terminal N-Boc aziridine 5 with
LTMP at -78 °C for 90 s, followed by the addition of CD3OD
as an external electrophile, did not lead to any of the expected
trans-deuterated N-Boc aziridine; instead, a 31% yield of N-H
aziridinylester 7 was isolated as a single diastereomer along
with 50% recovered starting material with 0% D incorporation.
The stereochemistry of ester 7 was initially assigned trans due
to previously observed trans-selective lithiation/silylation by the
sterically demanding LTMP with terminal epoxides and N-Bus
aziridines,19 and this assignment was later supported by crystal-
lographic analysis of a related aziridinylester prepared by similar
chemistry (vide infra).
SCHEME 2. In Situ Lithiation-Silylation13
It appeared that a N- to C-[1,2] lithiation-induced shift had
occurred resulting in synthetically valuable aziridinylester func-
tionality,20 along with concomitant N-deprotection. Although
lithiation-induced N- to C-1,2-shifts are known,9,21 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.21e In
this latter work, the N-Boc aziridine of styrene was treated with
s-BuLi in THF at -98 °C to give a phenyl-stabilized R-lithiated
aziridine that underwent migration to give 2-phenyl-2-Boc
aziridine (90%). As this migration is only mentioned as an unde-
sired byproduct, albeit in excellent yield, we considered that
the [1,2] shift of an R-lithiated terminal N-Boc aziridine was of
sufficient synthetic novelty and potential to investigate further.
Since terminal N-Boc aziridines are simple to access in an enan-
tiopure manner in two steps from a racemic terminal epoxide,2d
the migration reaction could potentially allow straightforward
access to important enantiopure N-H aziridinylesters.20
The first necessity was to improve the yield of the [1,2] shift
reaction. As only a mixture of aziridinylester 7 and starting
material 5 had been returned from leaving the reaction for 90
s, then initially the reaction time was investigated (Table 1).
Simply by increasing the reaction duration prior to quenching
led to an excellent 90% yield of aziridinylester 7 after 90 min
(entries 1-4). The reactions were remarkably clean with no
In this paper we report in detail our studies which significantly
expand the area of R-lithiated terminal aziridine chemistry. The
isomerization of R-lithiated N-Boc aziridines to N-H trans-
aziridinylesters,14 the dimerization of R-lithiated N-Bus aziri-
dines to 2-ene-1,4-diamines,15 and the intramolecular cyclopro-
panation of R-lithiated unsaturated N-Bus aziridines to trans-
2-aminobicyclo[3.1.0]hexanes16 is described.
Results and Discussion
Seeking to extend the chemistry of Beak and co-workers
(Scheme 2), we considered whether our previously developed
protocol for the diastereoselective in situ lithiation/silylation of
(5) Boche, G.; Lohrenz, J. C. W. Chem. ReV. 2001, 101, 697-756.
(6) (a) Arjona, O.; Menchaca, R.; Plumet, J. Heterocycles 2001, 55, 5-7.
(b) Mu¨ller, P.; Nury, P. HelV. Chim. Acta 2001, 84, 662-677. (c) O’Brien,
P.; Rosser, C. M.; Caine, D. Tetrahedron Lett. 2003, 44, 6613-6615. (d)
Mu¨ller, P.; Riegert, D.; Bernardinelli, G. HelV. Chim. Acta 2004, 87, 227-
239.
(7) (a) Moore, S. P.; Coote, S. C.; O’Brien, P.; Gilday, J. Org. Lett.
2006, 8, 5145-5148. (b) Hodgson, D. M.; Sˇtefane, B.; Miles, T. J.;
Witherington, J. J. Org. Chem. 2006, 71, 8510-8515.
(8) Huang, J.; O’Brien, P. Chem. Commun. 2005, 5696-5698.
(9) Vedejs, E.; Moss, W. O. J. Am. Chem. Soc. 1993, 115, 1607-1608.
(10) (a) Vedejs, E.; Kendall, J. T. J. Am. Chem. Soc. 1997, 119, 6941-
6942. (b) Vedejs, E.; Bhanu Prasad, A. S.; Kendall, J. T.; Russel, J. S.
Tetrahedron 2003, 59, 9849-9856. (c) Concello´n, J. M.; Ramo´n Sua´rez,
J.; Garc´ıa-Granda, S.; D´ıaz, M. R. Angew. Chem., Int. Ed. 2004, 43, 4333-
4336. (d) Concello´n, J. M.; Bernad, P. L.; Ramo´n Sua´rez, J. Chem. Eur. J.
2005, 11, 4492-4501.
(17) (a) Hodgson, D. M.; Reynolds, N. J.; Coote, S. J. Tetrahedron Lett.
2002, 43, 7895-7897. (b) Hodgson, D. M.; Kirton, E. H. M.; Miles, S.
M.; Norsikian, S. L. M.; Reynolds, N. J.; Coote, S. J. Org. Biomol. Chem.
2005, 3, 1893-1904.
(18) SÄliwin´ska, A.; Zwierzak, A. Tetrahedron 2003, 59, 5927-5934.
(19) Hodgson, D. M.; Humphreys, P. G.; Ward, J. G. Org. Lett. 2005,
7, 1153-1156.
(20) (a) Zwanenburg, B.; ten Holte, P. Top. Curr. Chem. 2001, 216, 93-
124. (b) Lee, W. K.; Ha, H.-J. Aldrichim. Acta 2003, 36, 57-63. (c) Zhou,
P.; Chen, B.-C; Davis, F. A. In Aziridines and Epoxides in Organic
Synthesis; Yudin, A. K., Ed.; Wiley-VCH: Weinheim, Germany, 2006; pp
73-115.
(21) (a) Vogel, C. Synthesis 1997, 497-505. (b) Kise, N.; Ozaki, H.;
Terui, H.; Ohya, K.; Ueda, N. Tetrahedron Lett. 2001, 42, 7637-7639. (c)
Kells, K. W.; Ncube, A.; Chong, J. M. Tetrahedron 2004, 60, 2247-2257.
(d) Dieltiens, N.; Stevens, C. V.; Masschelein, K. G. R.; Rammeloo, T.
Tetrahedron 2005, 61, 6749-6756. (e) Capriati, V.; Florio, S.; Luisi, R.;
Musio, B. Org. Lett. 2005, 7, 3749-3752.
(11) During manuscript preparation, an example of terminal aziridine
R-lithiation was reported, see: Luisi, R.; Capriati, V.; Di Cunto, P.; Florio,
S.; Mansueto, R. Org. Lett. 2007, 9, 3295-3298.
(12) Ratio not reported, in our hands a 1.7:1 trans/cis ratio was obtained.
(13) Beak, P.; Wu, S.; Yum, E. K.; Jun, Y. M. J. Org. Chem. 1994, 59,
276-277.
(14) Hodgson, D. M.; Humphreys, P. G.; Xu, Z.; Ward, J. G. Angew.
Chem., Int. Ed. 2007, 46, 2245-2248.
(15) Hodgson, D. M.; Miles, S. M. Angew. Chem., Int. Ed. 2006, 45,
935-938.
(16) Hodgson, D. M.; Humphreys, P. G.; Ward, J. G. Org. Lett. 2006,
8, 995-998.
10010 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
SCHEME 3. Terminal N-Boc Aziridine Synthesis and Lithiation-Electrophile Trapping
TABLE 1. Optimization of Lithiation-Induced Migration
aziridinylester 7, although increasing the temperature to ambient
temperature resulted in decomposition (entries 14 and 15).
Running the reaction at 0 °C throughout is also possible, though
a decreased yield (60%) was observed (entry 16). Finally, direct
application of the conditions of Florio and co-workers21e gave
a reduced 56% yield of aziridinylester 7, along with several
unidentified byproducts (entry 17).
With conditions that gave access to trans-aziridinylester 7
in 90% yield (entry 4), the scope of the reaction was next investi-
gated with a variety of terminal N-Boc aziridines (Table 2).
base
temp
recvd
yield
entry
(equiv)
solvent
(°C)
time
90 s
5 (%) of 7 (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
LTMP (3)
LTMP (3)
LTMP (3)
LTMP (3)
LTMP (3)
LTMP (3)
LTMP (3)
LDA (3)
THF
THF
THF
THF
Et2O
pentane
t-BuOMe -78
THF
THF
-78
50a
42a
2a
0
31
37
88
90
0
24
34
0
0
trace
5
50
64
74
0
60
56
-78
-78
-78
-78
-78
5 min
60 min
90 min
90 min
90 min
90 min
90 min
90 min
12 h
98a
10
12
98
98
b
TABLE 2. Scope of Lithiation-Induced Migration Reaction
-78
-78
-78
-78
-78
-78
LiNCy2 (3)
LTMP (0.1) THF
LTMP (0.5) THF
LTMP (1.1) THF
LTMP (2)
LTMP (3)
LTMP (3)
LTMP (3)
12 h
90 min
90 min
86
49
26
0
0
0
THF
THF
THF
THF
-78 to 0 90 min
-78 to rt 90 min
0
90 min
5 min
s-BuLi (1.2) THF
-98
0
a 0% D incorporation observed following quenching reaction with
CD3OD. b Not determined.
evidence of any 2-ene-1,4-diamine (as a result of carbene
dimerization, vide infra) observed by careful analysis of the
1
crude reaction mixture by TLC and H NMR spectroscopy.
Changing the solvent to diethyl ether resulted in shutting
down the migration reaction completely (entry 5): quenching
the reaction in Et2O after 90 min with CD3OD resulted in 0%
D incorporation into aziridine 5, implying that no lithiation was
occurring at -78 °C (entry 5). Changing the solvent to pentane
or t-BuOMe did result in small amounts of the desired
aziridinylester 7, but considerable amounts of unidentified
byproducts were also returned from these reactions (entries 6
22
and 7). Using the less basic lithium amides LDA or LiNCy2
resulted in no migration occurring (entries 8 and 9). As treatment
of the N-Boc aziridine would generate an aziridine lithium amide
(following [1,2] shift), it was considered whether catalytic
LTMP could be used in the reaction, with the aziridine lithium
amide regenerating LTMP from TMP. However, substoichio-
metric amounts of LTMP only resulted in poor conversions,
indicating that the reaction is not catalytic in LTMP (entries 10
and 11). Using 1.1 or 2 equiv of LTMP also resulted in poor
conversion to the product, with starting aziridine 5 recovered
(entries 12 and 13). Allowing the temperature of the reaction
to rise from -78 °C to 0 °C gave a 74% yield of the desired
a Reaction warmed from -78 °C to 0 °C for 90 min.
Tethered olefin and aromatic functionality were tolerated with
potentially competitive allylic deprotonation,23 cyclopropana-
J. Org. Chem, Vol. 72, No. 26, 2007 10011

Hodgson et al.
tion,16 or benzylic deprotonation all avoided under the reaction
conditions (Table 2, entries 1 and 2). X-ray crystallographic
analysis of aziridinylester 9b supported the assigned trans-
stereochemistry.24 TBS protected alcohols were tolerated (entries
3 and 4),25 along with a potentially eliminable/displaceable
primary chloride (entry 5). Using 2,2-disubstituted aziridine 8f2a
only returned starting material at -78 °C; however, on warming
to 0 °C a 67% yield of the desired aziridinylester was isolated
(entry 6). No degradation of ee was observed under the reaction
conditions (entry 7),24 though the reaction was not extendable
to 2,3-disubtituted aziridines: attempted reaction of aziridine
8g26 returned starting material at -78 °C (entry 8), presumably
due to unfavorable steric interactions. Warming the reaction
from -78 °C to 0 °C gave a mixture of starting material and
decomposition products and, consistent with results observed
with aziridine 5 (Table 1, entry 15), only decomposition was
observed upon warming to room temperature.
Due to the rapidity of carbenoid dimerization of R-lithiated
N-Bus terminal aziridines (vide infra), we sought to ascertain
whether the N- to C-migration reaction was inter- or intramo-
lecular in nature. A crossover experiment was designed21c
whereby different N-protecting groups would allow these two
processes to be distinguished (Scheme 4). Reaction of a 1:1
mixture of N-Boc aziridine 8e and N-tert-amylcarboxy aziridine
10 under the optimized migration conditions only gave a mixture
of aziridinylesters 9e and 11.27 No evidence of crossover of
either of the protecting groups was observed following careful
analysis of the crude reaction mixture and purified products,
indicating the migration reaction is intramolecular.
migration is the rate-limiting step.21a In the current reaction,
the [1,2] shift is complete within 90 min at -78 °C and lithiation
by LTMP appears to be the rate-limiting step. While a solvent-
caged (homolytic) cleavage-recombination mechanism cannot
be ruled out at this time, these results suggest that the reaction
could alternatively proceed via intramolecular attack of the
R-lithiated aziridine at the carbamoyl carbonyl.28 As simple
R-lithiated N-Boc heterocycles (5-8 membered rings) do not
display this rapid propensity of protecting group migration,13,29
then in the present case, migration may be assisted by a
comparatively reduced N lone pair/σ* CdO overlap due to the
significant increase in aziridine ring strain that such an overlap
would introduce.30 An indication of this reduced overlap can
be seen in the NMR spectra of terminal N-Boc aziridines where
rotameric signals typical of N-Boc amines are not observed.24
The increase of carbonyl stretching frequency in the IR spectrum
from ∼1690 cm-1 for typical secondary N-Boc amines to ∼1720
cm-1 for N-Boc aziridines also indicates a decreased level of
amide character in the latter case.
tert-Butyl protected N-H aziridinylesters accessed by this
methodology can undergo a variety of synthetically useful
subsequent transformations, such as those shown in Scheme 5.
Completely regioselective hydrogenolysis31 of aziridinylester
9a gave potentially useful32 protected â-amino acid 12. No
recourse to chromatography was necessary, with pure product
isolated following simple filtration and aqueous workup. Oxida-
tive cycloamination of the tethered olefin of 9a with N-bromo-
succinimide following the procedure of Yudin and co-workers30
gave azabicycle 13 as a separable 3.4:1 mixture of diastereomers.
Attempted elimination with KOH in methanolic THF resulted
in decomposition;30 however, use of DBU in toluene gave
enamine 14 in an excellent overall yield. Such bicyclic enamines
are known to undergo regioselective ring-opening to give
substituted pyrrolidines.30 With a view to achieving a synthesis
of the azirine containing natural product azirinomycin (vide
infra),33 regioselective Swern oxidation of aziridinylester 9a was
carried out according to the procedure of Zwanenburg and co-
workers.34 Synthetically useful35 azirine 15 was isolated as a
single regioisomer in 92% yield, setting the stage for a short
asymmetric synthesis of the tert-butyl ester of the unstable
natural product antibiotic azirinomycin (16)33 (Scheme 6).
Treatment of racemic propylene oxide (17) with 0.45 equiv
of tert-butyl carbamate under the aminolytic kinetic resolution
SCHEME 4. Crossover Experiment
As part of the optimization investigation of the lithiation-
induced [1,2]-shift reaction, the length of time the reaction
needed to progress to completion at -78 °C was studied (Table
1, entries 1-4). Following quenching these reactions after 90
s, 5 min, 60 min, and 90 min with CD3OD and careful analysis
(26) Mordini, A.; Russo, F.; Valacchi, M.; Zani, L.; Degl’Innocenti, A.;
Reginato, G. Tetrahedron 2002, 58, 7153-7163.
(27) Reaction of aziridine 10 alone gave aziridinylester 11 in 86% yield.
(28) (a) Eisch, J. J.; Kovacs, C. A. J. Organomet. Chem. 1971, 30, C97-
C100. (b) Eisch, J. J.; Dua, S. K.; Kovacs, C. A. J. Org. Chem. 1987, 52,
4437-4444.
(29) Gawley, R. E.; O’Connor, S.; Klein, R. In Science of Synthesis:
Houben-Weyl Methods of Molecular Transformations; Snieckus, V., Ed.;
Thieme: Stuttgart, Germany, 2005; Vol. 8a, pp 677-758.
(30) Chen, G.; Sasaki, M.; Li, X.; Yudin, A. K. J. Org. Chem. 2006, 71,
6067-6073.
1
of the H NMR spectra, no D incorporation was observed in
the starting material in any of the reactions. These results suggest
that once lithiation has occurred, the migration is rapid. The
length of reaction time required for consumption of the starting
material (90 min) indicates that slow lithiation by LTMP is the
rate-limiting step. Cleavage-recombination mechanisms have
been proposed for lithiation-induced N- to C-[1,2] shifts, though
these typically require warmer (>-78 °C) temperatures and
(31) Davis, F. A.; Deng, J.; Zhang, Y.; Haltiwanger, R. C. Tetrahedron
2002, 58, 7135-7143.
(32) EnantioselectiVe Synthesis of â-Amino Acids; 2nd ed.; Juaristi, E.,
Soloshonok, V. A., Eds.; Wiley: New York, 2005.
(22) Fraser, R. R.; Mansour, T. S. J. Org. Chem. 1984, 49, 3442-3443.
(23) Mordini, A.; Peruzzi, D.; Russo, F.; Valacchi, M.; Reginato, G.;
Brandi, A. Tetrahedron 2005, 61, 3349-3360.
(33) (a) Stapley, E. O.; Hendlin, D.; Jackson, M.; Miller, A. K.;
Hernandez, S.; Mata, J. M. J. Antibiot. 1971, 24, 42-47. (b) Miller, T. W.;
Tristram, E. W.; Wolf, F. J. J. Antibiot. 1971, 24, 48-50.
(34) Gentilucci, L.; Grijzen, Y.; Thijs, L.; Zwanenburg, B. Tetrahedron
Lett. 1995, 36, 4665-4668.
(35) (a) Palacios, F.; Ochoa de Retana, A. M.; Martinez de Marigorta,
E.; Manuel de los Santos, J. Eur. J. Org. Chem. 2001, 2401-2414. (b)
Palacios, F.; Ochoa de Retana, A. M.; Martinez de Marigorta, E.; Manuel
de los Santos, J. Org. Prep. Proced. Int. 2002, 34, 219-269.
(24) See the Supporting Information for details.
(25) TBS protected alcohol 9c has the potential to serve as a useful
intermediate in the synthesis of the aziridine-containing antibiotic (2S,3S)-
dicarboxyaziridine, isolated from a Streptomyces strain in 1975, see:
Naganawa, H.; Usui, N.; Takita, T.; Hamada, M.; Umezwa, H. J. Antiobiot.
1975, 28, 828-829.
10012 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
SCHEME 5. Aziridinylester Transformations
SCHEME 6. Synthesis of (S)-Azirinomycin Ester
conditions developed by Bartoli and co-workers2d gave amino
alcohol 18 in quantitative yield (based on tert-butyl carbamate)
and >99% ee.24 Alcohol tosylation, followed by KOH-mediated
aziridine ring-closure gave volatile aziridine (R)-3 in a single
flask transformation.2a Treatment with LTMP under the opti-
mized migration conditions gave the aziridinylester as a 19:1
trans/cis mixture that was purified to give trans-aziridinylester
19 in 70% yield. Presumably the formation of traces of the cis-
diastereomer is due to the reduced steric demand of the aziridine
methyl substituent. Finally, Swern oxidation with the previously
successful conditions (cf. Scheme 5) gave the tert-butyl ester
of natural (S)-azirinomycin 20 as a single regioisomer in only
4 steps from racemic propylene oxide.
epoxides (3 equiv of LTMP, 3 equiv of Me3SiCl, THF, 0 °C,
16 h) with terminal N-Bus aziridine 22a (Scheme 7).
SCHEME 7. Silylation or Dimerization of Terminal N-Bus
Aziridine 22a
Dimerization. While the above rapid N- to C-migration in
R-lithiated terminal N-Boc aziridines prevented electrophile
trapping at carbon (apart from Me3SiCl in situ, Scheme 3), we
have previously found this can be overcome by using N-Bus36
(tert-butylsulfonyl) protection. In the latter chemistry, provided
the electrophile is added relatively quickly (90 s after addition
to a solution of LTMP) then access to a wide variety of trans
R,â-substituted N-Bus aziridines was found to be possible.19
During our initial studies on the lithiation of terminal N-Bus
aziridines using in situ silylation, we investigated our previously
developed conditions17 for the in situ silylation of terminal
R,â-Aziridinylsilane 23 was isolated as a single trans-
diastereomer, and no evidence of N- to C-sulfonyl migration,
insertion by LTMP to give an enamine/aldehyde,37 or 2H-
azirine38/amino aldehyde formation (from hydrolysis of the
azirine) was observed, following careful analysis of the crude
(36) (a) Sun, P.; Weinreb, S. M.; Shang, M. J. J. Org. Chem. 1997, 62,
8604-8608. (b) Gontcharov, A. V.; Liu, H.; Sharpless, K. B. Org. Lett.
1999, 1, 783-786.
(37) Hodgson, D. M.; Bray, C. D.; Kindon, N. D. J. Am. Chem. Soc.
2004, 126, 6870-6871.
J. Org. Chem, Vol. 72, No. 26, 2007 10013

Hodgson et al.
TABLE 3. Dimerization of Terminal N-Bus Aziridines
a Mixture of three diastereomers isolated. b Inseparable mixture of olefin isomers in 93:7 ratio. c Reaction quenched after 10 min at -78 °C to give an
inseparable mixture of olefin isomers in a 57:43 ratio.
reaction mixture. Although this silylation was later optimized
these optimization studies, switching the solvent to hexane, Et2O,
or t-BuOMe did not lead to formation of the expected R,â-
to 86% by running the reaction at -78 °C for 1 h,19 during
10014 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
aziridinylsilane 23; instead a 60-63% yield of a mixture of
three distinct diastereomers of 2-ene-1,4-diamine 24a was
isolated (Scheme 7). The formation of diamine 24a was not
totally unexpected as analogous terminal epoxides undergo rapid
carbenoid eliminative dimerization at -5 °C to give 2-ene-1,4-
diols upon treatment with LTMP.39 Although the dimerization
reaction of an N-alkyl lithiated methylene aziridine (to give a
cyclopentene) had been demonstrated previously,40 the current
observation represented the first example of the carbenoid nature
of R-lithiated terminal aziridines, something that was of interest
given the potential for 2H-azirine formation via elimination of
the Bus group.38
2-Ene-1,4-diamines and their saturated derivatives are of
significant interest, as they have been used as chiral ligands in
catalysis and could serve as potential precursors for biologically
active HIV protease inhibitors.41-43 The current reaction pa-
rameters were investigated, and it was found that the presence
of the potentially Lewis acidic chlorotrimethylsilane was not
necessary for the dimerization reaction to proceed. The yield
of diamine 24a could be increased to 90% by increasing the
concentration (from 0.1 to 0.4 M in aziridine 22a), using a
solvent mixture of THF/hexanes (2:3) and warming the reaction
from -78 °C to 0 °C over 80 min (Table 3, entry 1).
As there are only two possible alkene geometries that could
arise from the dimerization of enantiopure terminal aziridines,
we anticipated that the reaction would have more potential for
the synthesis of symmetrical enantiopure 2-ene-1,4-diamines
from enantiopure terminal aziridines. Although there are some
excellent aminolytic kinetic resolution routes to access enan-
tiopure N-Boc, N-Ns, and N-SES terminal aziridines,2c,d we
envisaged a simple, robust three-step epoxide ring-opening/
aziridine ring-closure sequence to allow access to enantiopure
terminal N-Bus aziridines (Scheme 8).
contrast to the epoxide dimerization reaction process, which
typically gives E/Z alkene mixtures of the 2-ene-1,4-diol
products.39 Importantly for asymmetric synthesis, both enanti-
omers of 22b could be readily obtained and underwent the
dimerization reaction in excellent yields to allow access to either
enantiomer of diamine 24b (entries 2 and 3). The E-alkene
stereochemistry was proven unambiguously by X-ray crystal-
lographic analysis of diamine (S,S)-24c (entry 4).24 Steric bulk
in the γ-position had little effect on the reaction, with tert-butyl
and cyclohexyl groups tolerated (entries 4 and 5). Tethered
olefin functionality gave no evidence of allylic deprotonation23
and/or cyclopropanation (vide infra) (entries 6 and 7). Benzylic
deprotonation was avoided (entries 8 and 9); this is significant
since the enantiopure diamine products 24g are of interest in
the context of HIV protease inhibitor synthesis (vide infra).42
Interestingly, subjection of the analogous epoxide, 2,3-ep-
oxypropylbenzene, to the dimerization conditions gave cinnamyl
alcohol as the major product with no evidence of the 2-ene-
1,4-diol observed. This suggests that an (N-Bus) terminal
aziridine is more activated to R-(ring)-deprotonation than a
terminal epoxide. A trityl-protected alcohol was tolerated, as
was an achiral 2,2-disubstitued aziridine and an unsubstituted
aziridine (entries 10-12). With the two latter examples, a
mixture of heterochiral and homochiral dimerization of the
R-lithiated aziridines is possible, which could be the origin of
the E/Z alkene mixture of diamine products. Attempted in-
tramolecular dimerization was unsuccessful, even under dilute
conditions, with mixtures of intermolecular dimerization and
starting material recovered (entries 13 and 14).
The highly selective formation of the E-alkene isomer from
enantiopure substrates is consistent with an initial diastereose-
lective trans-lithiation,19 followed by nucleophilic attack of one
R-lithiated aziridine onto another acting as an electrophile; the
latter could occur via a 1,2-metallate shift process (Scheme 9).46
Subsequent syn elimination driven by the steric demand of the
aziridine substituents and the N-Bus groups leads to the observed
E-alkenes. The improved E/Z selectivity compared to the
analogous epoxide dimerization reactions is attributed to the
extra steric demand brought about by the presence of the Bus
SCHEME 8. Synthesis of Enantiopure N-Bus Aziridines
(42) (a) Lam, P. Y. S.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C. N.;
Ru, Y.; Bacheler, L. T.; Meek, J. L.; Otto, M. J.; Rayner, M. M.; Wong, Y.
N.; Chang, C.-H.; Weber, P. C.; Jackson, D. A.; Sharpe, T. R.; Erickson-
Viitanen, S. Science 1994, 263, 380-384. (b) Pierce, M. E.; Harris, G. D.;
Islam, Q.; Radesca, L. A.; Storace, L.; Waltermire, R. E.; Wat, E.; Jadhav,
P. K.; Emmet, G. C. J. Org. Chem. 1996, 61, 444-450. (c) Nugiel, D. A.;
Jacobs, K.; Worley, T.; Patel, M.; Kaltenbach, R. F., III; Meyer, D. T.;
Jadhav, P. K.; De Lucca, G. V.; Smyser, T. E.; Klabe, R. M.; Bacheler, L.
T.; Rayner, M. M.; Seitz, S. P. J. Med. Chem. 1996, 39, 2156-2169. (d)
Lam, P. Y. S.; Ru, Y.; Jadhav, P. K.; Aldrich, P. E.; DeLucca, G. V.;
Eyermann, C. J.; Chang, C.-H.; Emmett, G.; Holler, E. R.; Daneker, W. F.;
Li, L.; Confalone, P. N.; McHugh, R. J.; Han, Q.; Li, R.; Markwalder, J.
A.; Seitz, S. P.; Sharpe, T. R.; Bacheler, L. T.; Rayner, M. M.; Klabe, R.
M.; Shum, L.; Winslow, D. L.; Kornhauser, D. M.; Jackson, D. A.; Erickson-
Viitanen, S.; Hodge, C. N. J. Med. Chem. 1996, 39, 3514-3525. (e) Davis,
A. M.; Teague, S. J.; Kleywegt, G. J. Angew. Chem., Int. Ed. 2003, 42,
2718-2736. (f) Loughlin, W. A.; Tyndall, J. D. A.; Glenn, M. P.; Fairlie,
D. P. Chem. ReV. 2004, 104, 6085-6117.
Regioselective ring-opening of an enantiopure terminal ep-
oxide (accessed by Jacobsen’s hydrolytic kinetic resolution)44
with tert-butylsulfonamide gave the corresponding N-Bus
1-amino-2-ol.45 Alcohol mesylation, followed by base-induced
ring-closure gave the enantiopure terminal N-Bus aziridine in
typically good yields over the three steps.2c With a straightfor-
ward route to access a variety of enantiopure terminal N-Bus
aziridines, the scope of the reaction was investigated (Table 3).
Crucially, treatment of enantiopure (R)-22b with the opti-
mized dimerization conditions gave a 97% yield of diamine
1
(R,R)-24b as a single E-alkene isomer (as determined by H
NMR analysis), with none of the Z-alkene detectable in either
the crude or purified products (Table 1, entry 2). This is in
(43) For an alternative (nonstereoselective) base-induced carbene-type
homocoupling to a 2-ene-1,4-diamine, see: Weber, B.; Schwerdtfeger, J.;
Fro¨hlich, R.; Go¨hrt, A.; Hoppe, D. Synthesis 1999, 1915-1924
(44) Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Hansen,
K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc.
2002, 124, 1307-1315.
(38) Luisi, R.; Capriati, V.; Florio, S.; Di Cunto, P.; Musio, B.
Tetrahedron 2005, 61, 3251-3260.
(39) Hodgson, D. M.; Bray, C. D.; Kindon, N. D. Org. Lett. 2005, 7,
2305-2308.
(40) (a) Quast, H.; Weise Ve´lez, C. A. Angew. Chem., Int. Ed. 1974,
13, 342-343. (b) Dimerization of N-unprotected aziridines under Lewis
acid catalysis to give aminoaziridines is known, see: Caiazzo, A.; Dalili,
S.; Yudin, A. K. Synlett 2003, 2198-2202.
(41) Fache, F.; Schulz, E.; Tommasino, M. L.; Lemaire, M. Chem. ReV.
2000, 100, 2159-2231.
(45) Albanese, D.; Landini, D.; Penso, M.; Petricci, S. Tetrahedron 1999,
55, 6387-6394.
(46) (a) Kocien´ski, P.; Barber, C. Pure Appl. Chem. 1990, 62, 1933-
1940. (b) Hodgson, D. M.; Fleming, M. J.; Stanway, S. J. J. Org. Chem.
2007, 72, 4763-4773.
J. Org. Chem, Vol. 72, No. 26, 2007 10015

Hodgson et al.
SCHEME 9. Potential Dimerization-Elimination Pathways
SCHEME 10. DMP 323 and Its Potential Formation from a 1,4-Diamine-2,3-diol
SCHEME 11. Applications of Enediamine (R,R)-24g to the Synthesis of HIV Protease Inhibitors and Chiral Ligands
SCHEME 12. Metathesis Reactions of Trienediamines
groups. At this time, the origin of reaction pathway dependence
on protecting group (1,2 N- to C- for Boc, dimerization for Bus)
is unclear, although it should be mentioned that R-lithiated
N-phosphonate aziridines also undergo the 1,2 rearrangement
pathway to give trans-R,â-aziridinylphosphonates.14
1,4-Diamine derivatives, particularly cyclic ureas such as the
DuPont Merck compound DMP 323 have received considerable
interest due to their use as peptidomimetic HIV protease inhib-
itors (Scheme 10).42 As mentioned above, a common motif of
these biologically active molecules is a 1,4-diamine-2,3-diol unit
with R,S,S,R configuration with 1,4-benzyl groups which could
conceivably be derived from enantiopure diamine (R,R)-24g.42b
To demonstrate the utility of the dimerization methodology,
enantiopure diamine (R,R)-24g was dihydroxylated by using
in excellent yield. This is not only the core asymmetric unit of
catalytic OsO4 with stoichiometric NMO to give 1,4-amine-
a number of highly active HIV protease inhibitors,42,48 but it
2,3-diol 25 in a 7:1 diastereomeric crude mixture that was
also has found use as a ligand or ligand precursor in a number
purified to give an 83% yield of (R,S,S,R)-25 (Scheme 11).
of enantioselective transformations.49 Derivatization of ent-26,
X-ray crystallographic analysis also proved the assigned ster-
which could be accessed similarly from (S,S)-24g (Table 3, entry
eochemistry.24 The dihydroxylation diastereoselectivity is in
9), has also previously been used to produce arrays of HIV
contrast to the 2:3 syn:anti selectivity previously observed with
protease inhibitors for biological evaluation, yielding a number
the analogous N-Boc 2-ene-1,4-diamine.47 Global Bus depro-
of very active compounds.50 Hydrogenation of diamine (R,R)-
tection was achieved upon treatment with triflic acid in the
24g under mild conditions was achieved over palladium on
presence of anisole36a to give the fully deprotected amine 26,
carbon with a balloon of hydrogen to give saturated diamine
27 in 99% yield. This latter reaction illustrates the utility of the
(47) (a) Rao, A. V. R.; Gurjar, M. K.; Pal, S.; Pariza, R. J.; Chorghade,
M. S. Tetrahedron Lett. 1995, 36, 2505-2508. (b) Gurjar, M. K.; Pal, S.;
Rao, A. V. R.; Pariza, R. J.; Chorghade, M. S. Tetrahedron 1997, 53, 4769-
(48) Dondoni, A.; Perrone, D.; Rinaldi, M. J. Org. Chem. 1998, 63,
9252-9264.
4778.
10016 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
SCHEME 13. r-Lithiated Aziridine Intramolecular Cyclopropanation
dimerization methodology to access enantiopure C2 symmetric
1,4-diamines which have found use as chiral ligands in
asymmetric transformations.41
As a further use of the diamine products, the metathesis
reactivity of 2-ene-1,4-diamines (S,S)-24e and (R,R)-24f bearing
further unsaturation was investigated. Interestingly, treatment
of (S,S)-24e with 5 mol % of Grubbs second generation catalyst
gave cyclopentene 28, whereas reaction of (R,R)-24f under
similar conditions gave 18-membered macrocycle 29 as a 9:1
mixture of olefin geometric isomers (Scheme 12).
Intramolecular Cyclopropanation. Following on from the
success of the N-Bus group in R-lithiated aziridine carbenoid
dimerization chemistry, we considered whether conditions could
be developed for an alternative carbenoid process, namely
intramolecular cyclopropanation to access 2-aminobicyclo[3.1.0]-
hexanes. This latter structural motif is found in a variety of
biologically and pharmaceutically interesting targets, such as
analgesics,51 potential anti-obesity therapeutics,52 anti-viral
agents,53 and anti-diabetic agents.54
of N-tosyl aziridine 302b,56 with 2 equiv of LTMP at 0 °C
followed by warming to room temperature for 16 h gave bicyclic
amine 31 as a single trans-diastereomer as judged by NOE
analysis,24 albeit in only 19% yield (Scheme 13). A screen of
different lithium amides revealed that the yield could be
increased to 37% by using the slightly less hindered lithium
dicyclohexylamide.24
We suspected that competitive ortho-lithiation of the tosyl
group was responsible for the reduced yield of bicyclic amine
31.8 As such, a screen was initiated with use of t-Bu, Boc,
mesitylsulfonyl, p-methoxybenzenesulfonyl, trisyl, and Bus as
alternative protecting groups either less likely, or unable to
undergo ortho-lithiation.24 All of the protecting groups inves-
tigated resulted in <15% yield of the bicyclic amine, apart from
N-Bus, which gave a 23% and 38% yield of a single diastere-
omer of bicyclic amine 32 with LTMP or lithium dicyclohexy-
lamide, respectively (Table 4, entries 1 and 2). The trans-
stereochemistry of bicyclic amine 32 was supported by X-ray
crystallographic analysis.24 As might be expected, considering
the rapidity of dimerization of aziridine 22e (cf. Table 3),
racemic 2-ene-1,4-diamine 24e was also isolated in 44% and
35% yield, respectively, as a mixture of three diastereomers. It
is noteworthy that cyclopropanation can occur at all at 0 °C,
given the propensity of R-lithiated N-Bus aziridines to undergo
competitive dimerization.
We began investigations using our previously developed
conditions for the intramolecular cyclopropanation of unsatur-
ated terminal epoxides to bicyclo[3.1.0]hexan-2-ols.55 Reaction
(49) (a) Zhang, A.; Feng, Y.; Jiang, B. Tetrahedron: Asymmetry 2000,
11, 3123-3130. (b) Zhao, J.; Bao, X.; Liu, X.; Wan, B.; Han, X.; Yang,
C.; Hang, J.; Feng, Y.; Jiang, B. Tetrahedron: Asymmetry 2000, 11, 3351-
3359. (c) Jiang, B.; Feng, Y.; Hang, J.-F. Tetrahedron: Asymmetry 2001,
12, 2323-2329.
TABLE 4. Optimization of the Synthesis of Bicyclic Amine 32
(50) (a) Kempf, D. J.; Codacovi, L.; Wang, X. C.; Kohlbrenner, W. E.;
Wideburg, N. E.; Saldivar, A.; Vasavanonda, S.; Marsh, K. C.; Bryant, P.;
Sham, H. L.; Green, B. E.; Betebenner, D. A.; Erickson, J.; Norbeck, D.
W. J. Med. Chem. 1993, 36, 320-330. (b) Budt, K.-H.; Peyman, A.; Hansen,
J.; Knolle, J.; Meichsner, C.; Paessens, A.; Ruppert, D.; Stowasser, B.
Bioorg. Med. Chem. 1995, 3, 559-571. (c) Martin, S. F.; Dorsey, G. O.;
Gane, T.; Hillier, M. C.; Kessler, H.; Baur, M.; Matha¨, B. J. Med. Chem.
yield (%)
concn
1998, 41, 1581-1597.
(51) Moshe, A. Eur. J. Med. Chem. 1981, 16, 199-206.
entrya
base
equiv
solvent
(M)
temp
32
24e
base added to aziridine 22e over 1 h
(52) (a) McBriar, M. D.; Guzik, H.; Xu, R.; Paruchova, J.; Li, S.; Palani,
A.; Clader, J. W.; Greenlee, W. J.; Hawes, B. E.; Kowalski, T. J.; O’Neill,
K.; Spar, B.; Weig, B. J. Med. Chem. 2005, 48, 2274-2277. (b) McBriar,
M. D.; Guzik, H.; Shapiro, S.; Paruchova, J.; Xu, R.; Palani, A.; Clader, J.
W.; Cox, K.; Greenlee, W. J.; Hawes, B. E.; Kowalski, T. J.; O’Neill, K.;
Spar, B. D.; Weig, B.; Weston, D. J.; Farley, C.; Cook, J. J. Med. Chem.
2006, 49, 2294-2310. (c) Kowalski, T. J.; Spar, B. D.; Weig, B.; Farley,
C.; Cook, J.; Ghibaudi, L.; Fried, S.; O’Neill, K.; Del Vecchio, R. A.;
McBriar, M.; Guzik, H.; Clader, J.; Hawes, B. E.; Hwa J. Eur. J. Pharm.
2006, 535, 182-191. (d) McBriar, M. D.; Guzik, H.; Shapiro, S.; Xu, R.;
Paruchova, J.; Clader, J. W.; O’Neill, K.; Hawes, B.; Sorota, S.; Margulis,
M.; Tucker, K.; Weston, D. J.; Cox, K. Bioorg. Chem. Med. Lett. 2006,
16, 4262-4265.
(53) (a) Rodriquez, J. B.; Marquez, V. E.; Nicklaus, M. C.; Mitsuya,
H.; Barchi, J. J., Jr. J. Med. Chem. 1994, 37, 3389-3399. (b) Marquez, V.
E.; Siddiqui, M. A.; Ezzitouni, A.; Russ, P.; Wang, J.; Wagner, R. W.;
Matteucci, M. D. J. Med. Chem. 1996, 39, 3379-3747. (c) Tchilibon, S.;
Joshi, B. V.; Kim, S.-K.; Duong, H. T.; Gao, Z.-G.; Jacobson, K. A. J.
Med. Chem. 2005, 48, 1745-1758. (d) Comin, M. J.; Agbaria, R.; Ben-
Kasus, T.; Hulaihel, M.; Liao, C.; Sun, G.; Nicklaus, M. C.; Deschamps, J.
R.; Parrish, D. A.; Marquez, V. E. J. Am. Chem. Soc. 2007, 129, 6216-
6222.
1
2
LTMP
LiNCy2
2
2
t-BuOMe
t-BuOMe
0.07
0.07
0 °C to rt 23
0 °C to rt 38
44
35
aziridine 22e added to base over 1 h
3
4
5
6
7
LTMP
2
2
2
2
2
2
3
3
3
3
3
3
3
3
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
THF
0.07
0.07
0.07
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0 °C to rt 41
0 °C to rt 54
23
6
6
<5
7
16
<5
10
9
36
<5
<5
<5
<5
LiNCy2
LiNCy2
LiNCy2
LiNCy2
LiNCy2
LiNCy2
LTMP
0 °C
0 °C
-10 °C
-40 °C
0 °C
0 °C
0 °C
0 °C
0 °C
65
68
67
67
73
69
63
29
71
68
72
75
8
9
10
11
12
13
14
15b
16c
LDA
LiNCy2
LiNCy2
LiNCy2
LiNCy2
LiNCy2
hexane
Et2O
t-BuOMe
t-BuOMe
0 °C
0 °C
0 °C
a 16 h reaction duration unless otherwise indicated. b 1 h reaction
duration. c 2 h reaction duration.
(54) Nandanan, E.; Jang, S.-Y.; Moro, S.; Kim, H. O.; Siddiqui, M. O.;
Russ, P.; Marquez, V. E.; Busson, R.; Herdewijn, P.; Harden, T. K.; Boyer,
J. L.; Jacobson, K. A. J. Med. Chem. 2000, 43, 829-842.
(55) (a) Hodgson, D. M.; Chung, Y. K.; Paris, J.-M. J. Am. Chem. Soc.
2004, 126, 8664-8665. (b) Hodgson, D. M.; Chung, Y. K.; Nuzzo, I.;
Freixas, G.; Kulikiewicz, K. K.; Cleator, E.; Paris, J.-M. J. Am. Chem. Soc.
2007, 129, 4456-4462.
As dimerization is an intermolecular process, it was consid-
ered that the reaction parameters might be able to be biased to
(56) Lapinsky, D. J.; Bergmeier, S. C. Tetrahedron 2002, 58, 7109-
7117.
J. Org. Chem, Vol. 72, No. 26, 2007 10017

Hodgson et al.
TABLE 5. Bicyclic Amines from Unsaturated Terminal Aziridines
favor the intramolecular cyclopropanation reaction pathway
(Table 4).
A drop in yield of diamine 24e and an increase in yield of
bicyclic amine 32 was seen by simply reversing the order of
addition, so that the aziridine was added to base over 1 h (entries
3 and 4). This trend was particularly noticeable when using
LiNCy2 as the base (entry 4). Maintaining the reaction temper-
ature at 0 °C led to an increase in yield of bicyclic amine 32
and levels of diamine 24e were reduced to trace amounts by
halving the reaction concentration (entries 5 and 6). Further
lowering the temperature gave increased amounts of diamine
24e (entries 7 and 8). Increasing the equivalents of LiNCy2 to
3 allowed a 73% yield of bicyclic amine 32 to be obtained (entry
9). Reinvestigation of alternative lithium amides (LTMP and
LDA) gave increased amounts of diamine 24e (entries 11 and
12). A solvent screen revealed that the reaction proceeded
smoothly in hexane or Et2O, but diamine 24e formation
increased significantly in THF (entries 12-14). The latter result
with THF was expected, as the optimal conditions for the
dimerization reaction use THF as the solvent (Table 3). Finally,
quenching the reaction directly after competition of addition
and following stirring for 1 h demonstrated complete conversion
to bicyclic amine 32 (entries 15 and 16). This is in direct contrast
to carrying out the reaction by adding LiNCy2 to aziridine 22e
and quenching following complete addition: aziridine 22e was
recovered in 52%, bicyclic amine 32 in 10%, and diamine 24e
in 22% yield.
These results demonstrate that aziridine deprotonation and
intramolecular cyclopropanation is rapid when the aziridine is
added to the base. The dependence on order of addition could
be related to the aggregated nature of lithium amides:57 under
the latter conditions deprotonation of more than one aziridine
is unlikely to occur from the same lithium amide aggregate. In
contrast, when the base is added to the aziridine, this may create
a localized high concentration of lithiated aziridine,58 which
allows dimerization to become competitive with intramolecular
cyclopropanation. The greater success with LiNCy2 may be
related to a slightly reduced pKa value relative to LTMP,59 which
allows for a more matched rate of deprotonation and cyclopro-
panation of aziridine 22e thus preventing formation of a high
R-lithiated aziridine concentration and so retarding the dimer-
ization pathway.
To examine the scope of the intramolecular cyclopropanation
reaction, a range of unsaturated terminal N-Bus aziridines 34a-p
were synthesized24 and subjected to the optimized conditions
(Table 5).
Cyclopropanation of an enantiopure terminal aziridine gave
chiral bicyclic amine (+)-34a with no loss of enantiopurity
(entry 1).24 The reaction tolerated di- and trisubstituted alkenes
(entries 2-5) with the olefin geometry maintained into the
product (entries 2 and 3), indicating the stereospecificity of the
process. The decreased yield of with Z-olefin 34c was attributed
to a reduced rate of cyclopropanation brought about by the ethyl
group residing in a pseudoaxial position in the chairlike
transition state (cf. Scheme 13). This was also supported by
1
increased levels of dimer signals in the H NMR of the crude
(57) (a) Aubrecht, K. B.; Collum, D. B. J. Org. Chem. 1996, 61, 8674-
8676. (b) Remenar, J. F.; Lucht, B. L.; Collum, D. B. J. Am. Chem. Soc.
1997, 119, 5567-5572.
(58) For an example of the effect of localized high concentrations of
organolithium reagents, see: Beak, P.; Musick, T. J.; Chen C. J. Am. Chem.
Soc. 1988, 110, 3538-3545.
(59) Fraser, R. R.; Mansour, T. S. J. Org. Chem. 1984, 49, 3442-3443.
10018 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
SCHEME 14. N-Bus Deprotection
reaction mixture. X-ray crystallographic analysis supported the
trans stereochemistry of bicyclic amine 34d.24 Substitution on
the tether was also tolerated (entries 6-9), with excellent
selection between diastereotopic allyl groups obtained (entries
8 and 9). A tricyclic amine could be accessed as a single
diastereomer and aziridinyl-substituted styrenes were tolerated
under the reaction conditions (entries 10-14). Efficient access
to 2-amino-5-aryl-substituted bicyclo[3.1.0]hexanes such as 34m
is of interest for the synthesis of the potential anti-obesity
therapeutic trans-SCH-A (vide infra) and analogues.52 Attempted
cyclopropanation of an aziridine containing a conjugated diene
was poor yielding, attributed to competing allylic deprotonation
(entry 14).55b Trishomoallylic aziridine 34o underwent cyclo-
propanation in low yield, which was attributed to a slower rate
of cyclopropanation relative to dimerization, indicated by
deprotection under mild conditions was considered. Cyclopro-
panation of aziridine 33h on an 8 mmol scale used slightly
modified conditions that avoided chromatography and demon-
strated the potential for (modest) scale-up (Scheme 14). Ben-
zylation was achieved by using BnBr and NaH62 which gave
protected secondary amine 35 in 90% yield. Subjecting protected
secondary amine 35 to an excess of trifluoroacetic acid and
anisole at room temperature for 12 h gave the desired N-Bn
bicyclic amine 36 in 72% yield, along with 10% of acidic
decomposition product N-Bus amine 37. Finally, hydrogenolysis
using a balloon of hydrogen over palladium on carbon gave
primary bicyclic amine 38, in 94% yield.
To demonstrate the utility of this intramolecular R-lithiated
aziridine cyclopropanation process, we focused on a synthesis
of the epimer of the melanin-concentrating hormone receptor 1
(MCH-R1) antagonist trans-SCH-A (39), a highly promising
potential anti-obesity therapeutic developed by Schering-Plough
(Figure 1).52
1
increased dimer signals in the H NMR spectrum of the crude
reaction mixture. Using the slower rate of cyclopropanation for
a trishomoallylic aziridine to our advantage, an aziridine
containing both bis- and trishomoallylic groups underwent
selective cyclopropanation to give an 85% yield of 2-amino-
bicyclo[3.1.0]hexane 34p containing a potentially useful vinyl-
cyclopropane moiety.60
With access to a variety of trans-2-aminobicyclo[3.1.0]-
hexanes, our attention turned to deprotection of the N-Bus group.
Given the ability of cyclopropanes to assist in stabilization of
adjacent carbocation formation,61 we were concerned that the
strongly acidic conditions required for N-Bus removal36a could
result in degradation via elimination of the sulfonamide. Weinreb
and co-workers have shown that N-Bus protected primary
amines undergo deprotection more slowly and require stronger
acidic conditions than analogous N-Bus protected secondary
amines.36a Initial attempts with bicyclic amine 34g using
conditions developed for the deprotection of N-Bus primary
amines (triflic acid/anisole) resulted in decomposition. Switching
to trifluoroacetic acid, typically used for the deprotection of
N-Bus secondary amines, only returned starting material, even
after extended reaction times of 48 h. As conditions suitable
for the removal of N-Bus groups from protected secondary
amines did not result in cyclopropane rupture, conversion of
bicyclic amine 34h to protected secondary amine 35 to allow
FIGURE 1. trans-SCH-A.
The synthesis commenced with Rosenmund-von Braun
coupling between known alkenone 4063 (accessed by homoal-
lylic Grignard addition to 4-bromobenzaldehyde, followed by
Swern oxidation)24 and CuCN (Scheme 15).64 Epoxide forma-
tion via the bromohydrin, then Wittig olefination gave the
desired bishomoallylic epoxide 42 in 55% yield over the three
steps. Conversion to the bishomoallylic aziridine 43 was simply
accomplished by the three-step protocol discussed previously
(cf. Scheme 8). Crucially, given the potential for ortho-lithiation
of the aromatic ring upon treatment with LiNCy2,65 cyclopro-
panation of aziridine 43 proceeded smoothly to give key bicyclic
amine 44, in 87% yield. Monoalkylation with the known
piperazine 4566 gave protected secondary amine 46 in 93% yield.
Deprotection with triflic acid and anisole at 0 °C for 90 s was
found to be optimal24 for N-Bus removal, with the desired amine
47 being isolated in 86% yield as a single trans-diastereomer
as judged by NOE analysis.24 Importantly, the potentially acid
labile nitrile group was carried unscathed through the depro-
(60) (a) Hudlicky, T.; Kutchan, T. M.; Naqvi, S. M. Org. React. 1985,
33, 247-335. (b) Hudlicky, T.; Reed, J. W. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Paquette, L. A., Eds.; Pergamon:
Oxford, UK, 1992; Vol. 5, pp 899-970. (c) Oxgaard, J.; Wiest, O. Eur. J.
Org. Chem. 2003, 1454-1462. (d) Zuo, G.; Louie, J. Angew. Chem., Int.
Ed. 2004, 43, 2277-2279.
(61) Kirby, A. J. Stereoelectronic Effects; Oxford University Press:
Oxford, UK, 1996; pp 44-46.
J. Org. Chem, Vol. 72, No. 26, 2007 10019

Hodgson et al.
SCHEME 15. Synthesis of cis-SCH-A
tection step. Finally, urea formation with the commercially
available isocyanate 48 completed the synthesis of cis-SCH-A
(49) in 93% yield (16% over 11 steps).
Biological assay of cis-SCH-A (49) in a melanin concentrat-
ing hormone receptor 1 binding assay demonstrated no inhibition
of MCH-R1,24 presumably indicating a crucial dependence on
efficacy of the anti geometry between the amine and aromatic
ring of trans-SCH-A (39).
In conclusion, we report the discovery and development of
three new reactions of R-lithiated terminal aziridines and their
application to synthetically and biologically important targets.
In seeking to extend the scope of in situ lithiation/silylation
reactions of terminal N-Boc aziridines, a novel mode of N- to
C-Boc migration was observed. In general, good yields and
excellent diastereoselectivities were obtained for the syntheti-
cally useful trans-R,â-aziridinylesters and their utility has been
demonstrated with the synthesis of a protected â-amino acid,
and an asymmetric synthesis of (S)-azirinomycin tert-butyl ester
(20) in four steps from racemic propylene oxide. On switching
aziridine protecting groups from alkoxycarbonyl to organosul-
fonyl, a change in R-lithiated terminal aziridine reactivity was
observed. Access to a wide range of enantiopure 2-ene-1,4-
diamines in good yields was achieved by the carbenoid
dimerization of enantiopure terminal N-Bus aziridines. The
utility of these diamines was demonstrated by the efficient and
selective synthesis of diaminodiol (R,S,S,R)-26, the core unit
of a number of potent HIV protease inhibitors. R-Lithiated
terminal N-Bus aziridines bearing unsaturation were shown to
undergo intramolecular carbenoid cyclopropanation to give
single trans-diastereomers of synthetically useful 2-amino-
bicyclo[3.1.0]hexanes in good yields. The utility of this latter
methodology was demonstrated in the diastereoselective syn-
thesis of cis-SCH-A (49), the epimer of a highly effective
potential anti-obesity therapeutic.
Experimental Section
General experimental details are described in the Supporting
Information.
General Procedure: Optimized Conditions for R,â-Aziridi-
nylester Formation via LTMP-Mediated Migration of Terminal
N-Boc Aziridines, Described for tert-Butyl (2R*,3S*)-3-Buty-
laziridine-2-carboxylate 7. n-BuLi (1.6 M in hexanes, 0.94 mL,
1.5 mmol) was added dropwise to a stirred solution of 2,2,6,6-
tetramethylpiperidine (0.25 mL, 1.5 mmol) in THF (3.8 mL) at
-78 °C under argon. Following warming to room temperature for
30 min, the resulting solution was re-cooled to -78 °C and a
solution of aziridine 5 (100 mg, 0.50 mmol) in THF (1.5 mL) was
added dropwise over 1 min. Following stirring for 90 min at -78
°C, sat. aq NH4Cl (2 mL) was added and the flask was warmed to
room temperature. The aqueous phase was washed with Et2O (3 ×
10 mL). The combined organic phase was dried (MgSO4) and then
evaporated under reduced pressure. Purification of the residue by
column chromatography (SiO2, petroleum ether/Et2O, 9:1) gave
aziridinylester 7 as a colorless oil (90 mg, 90%). Rf 0.15 (petroleum
ether/Et2O 9:1); IR (neat) 3287 m (N-H), 2933 s, 861 s, 1721 s
(CdO), 1459 m, 1429 m, 1393 s, 1229 s, 1164 s, and 1029 m
(62) Bergeron, R. J.; Neims, A. H.; McManis, J. S.; Hawthorne, T. R.;
Vinson, J. R. T.; Bortell, R.; Ingeno, M. J. J. Med. Chem. 1988, 31, 1183-
1190.
(63) Hok, S.; Schore, N. E. J. Org. Chem. 2006, 71, 1736-1738.
(64) Friedman, L.; Shechter, H. J. Org. Chem. 1961, 26, 2522-2524.
(65) Fraser, R. R.; Bresse, M.; Mansour, T. S. J. Am. Chem. Soc. 1983,
105, 7790-7791.
(66) Barrett, P. A.; Caldwell, A. G.; Walls, L. P. J. Chem. Soc. 1961,
2404-2418.
1
cm-1; H NMR (400 MHz, CDCl3) δ 2.16-2.11 (2H, m), 1.46-
10020 J. Org. Chem., Vol. 72, No. 26, 2007

R-Lithiated Terminal Aziridines
1.33 (15H, m), 1.23-1.18 (1H, m), 0.89 (3H, t, J ) 7 Hz); 13C
NMR (100 MHz, CDCl3) δ 171.8 (C), 81.7 (C), 39.1 (CH), 36.1
(CH), 32.3 (CH2), 29.2 (CH2), 28.0 (3 × CH3), 22.3 (CH2), 14.0
(CH3); MS CI m/z (rel intensity) 200 (M + H+, 5), 144 (100), 100
(10), 98 (15); HRMS m/z calcd for C11H22NO2 200.1651, found
200.1656.
30 min, the reaction was cooled to 0 °C and a solution of aziridine
22e (109 mg, 0.5 mmol) in t-BuOMe (10 mL) at 0 °C was added
dropwise over 1 h via cannula and the reaction was left to stir at 0
°C. On completion of the reaction (2 h, tlc monitoring), sat. aq
NH4Cl (5 mL) and Et2O (5 mL) were added. The phases were
separated and the aqueous layer extracted with Et2O (2 × 15 mL).
The organic layers were combined, dried (MgSO4), and evaporated
under reduced pressure. Purification of the residue by column
chromatography (SiO2, petroleum ether/Et2O, 4:1) gave bicyclic
amine 32 as a white solid (82 mg, 75%). Mp 101-102 °C; Rf 0.38
(petroleum ether/Et2O 4:1); IR (film) 3273 s (N-H), 3036 m
(cyclopropane), 2935 s, 1454 m, 1364 w (SdO), 1297 s, 1131 s
General Procedure: Optimized Conditions for 2-Ene-1,4-
Diamine Formation via LTMP-Mediated Dimerization of Enan-
tiopure Terminal N-Bus Aziridines, Described for (12E,11R,-
14R)-N,N′-Bis(tert-butylsulfonyl)tetracos-12-ene-11,14-
diamine (R,R)-24b. n-BuLi (1.6 M in hexanes, 0.24 mL, 0.39
mmol) was added dropwise to a stirred solution of 2,2,6,6-
tetramethylpiperidine (66 µL, 0.39 mmol) in THF (0.4 mL) at -78
°C. The mixture was warmed to 0 °C over 15 min, then re-cooled
to -78 °C before dropwise addition of aziridine (R)-22b (38 mg,
0.13 mmol) in THF (0.8 mL). The mixture was stirred at -78 °C
for 20 min, then at 0 °C for 1 h, before the addition of MeOH (0.8
mL), sat. aq NH4Cl (8 mL), and Et2O (16 mL). The layers were
separated, and the aqueous phase was extracted with Et2O (16 mL).
The combined organic phase was dried (MgSO4) and concentrated.
Purification of the residue by column chromatography (SiO2,
petroleum ether/Et2O, 3:7) gave 2-ene-1,4-diamine (R,R)-24b as a
1
(SdO), 1108 m, 1058 w, and 1025 m cm-1; H NMR (400 MHz,
CDCl3) δ 4.16 (1H, d, J ) 9 Hz), 3.88-3.84 (1H, m), 1.89-1.79
(1H, m), 1.73-1.63 (2H, m), 1.52-1.46 (1H, m), 1.44-1.41 (2H,
m), 1.40 (9H, s), 0.48-0.42 (1H, m), 0.13-0.10 (1H, m); 13C NMR-
(100 MHz, CDCl3) δ 59.6 (C), 57.6 (CH), 29.6 (CH2), 24.4 (CH2),
24.3 (3 × CH3), 23.1 (CH), 16.4 (CH), 7.0 (CH2); MS CI m/z (rel
intensity) 235 (M + NH4+, 45), 218 (M + H+, 30), 98 (100), 81
(50); HRMS m/z calcd for C10H20NO2S 218.1215, found 218.1209.
Acknowledgment. We thank the EPSRC and GlaxoSmith-
Kline for a CASE award (to P.G.H.) and the EPSRC for a
Research Grant (GR/S46789/01). We thank Dr. B. Hawes
(Schering Plough) for running biological assays, Dr. G. Mac-
donald for useful discussions, and S. P. Hughes for preliminary
investigations. We also thank the EPSRC National Mass
Spectrometry Service (Swansea) for mass spectra and Prof. W.
Clegg (Newcastle), Dr. L. Russo (Newcastle), Dr. A. Cowley
(Oxford), and Dr. A. Collins (Oxford) for X-ray crystallographic
analyses.
colorless oil (37 mg, 97%). [R]25 -9.2 (c 2.0, CHCl3); Rf 0.50
D
(petroleum ether/Et2O 3:7); IR (neat) 3277 br s, 2957 s, 2925 s,
2855 s, 1633 w, 1480 s, 1457 s, 1433 s, 1366 m, 1306 s, 1263 m,
1
1126 s cm-1; H NMR (500 MHz, CDCl3) δ 5.59-5.55 (2H, m),
4.01-3.90 (4H, m), 1.65-1.49 (4H, m), 1.37 (18H, s), 1.33-1.24
(32H, m), 0.87 (6H, t, J ) 7 Hz); 13C NMR (125 MHz, CDCl3) δ
132.2 (2 × CH), 59.7 (2 × C), 56.4 (2 × CH), 37.2 (2 × CH2),
31.8 (2 × CH2), 29.5 (2 × CH2), 29.5 (2 × CH2), 29.5 (2 × CH2),
29.4 (2 × CH2), 29.2 (2 × CH2), 25.6 (2 × CH2), 24.2 (6 × CH3),
22.6 (2 × CH2), 14.0 (2 × CH3); MS CI m/z (rel intensity) 624 (M
+ NH4+, 15), 489 (15), 472 (30), 352 (80), 170 (100); HRMS m/z
calcd for C32H70N3O4S2 requires 624.4808, found 624.4839.
General Procedure: Optimized Conditions for trans-2-
Aminobicyclo[3.1.0]hexane Formation via LiNCy2-Mediated
Cyclopropanation of Bis-Homoallylic Terminal N-Bus Aziri-
dines, Described for N-[(1R*,2R*,5S*)-Bicyclo[3.1.0]hexan-2-
yl]-2-methylpropane-2-sulfonamide 32. n-BuLi (1.6 M in hexanes,
0.94 mL, 1.5 mmol) was added dropwise to a stirred solution of
dicyclohexylamine (0.30 mL, 1.5 mmol) in t-BuOMe (5 mL) at
-78 °C under argon. Following warming to room temperature for
Supporting Information Available: Full experimental details
of the syntheses and characterization of aziridine substrates, 2-ene-
1,4-diamines, bicyclic amines, and aziridinylesters not described
in the Experimental Section, X-ray data for aziridinylester 9b,
aziridine 22j, diamines (S,S)-24c and 25, and bicyclic amines 32
and 34d in CIF format, and copies of 1H and 13C NMR spectra for
all compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO701901T
J. Org. Chem, Vol. 72, No. 26, 2007 10021