Amphoteric amino aldehydes enable rapid assembly of unprotected amino alcohols

Article abstract of DOI:10.1002/anie.200705776

(Chemical Equation Presented) Apples and oranges: The term "amphoteric" is derived from the Greek "amphoteros", which literally means "both of two". Amphoteric amino aldehydes are counterintuitive molecules in that they contain both electrophilic and nucl

Full text of DOI:10.1002/anie.200705776

Communications
DOI: 10.1002/anie.200705776
Synthetic Methods
Amphoteric Amino Aldehydes Enable Rapid Assembly of
Unprotected Amino Alcohols**
Ryan Hili and Andrei K. Yudin*
A variety of molecules that contain functionally significant b-
amino alcohol motifs[1] are assembled by the addition of
carbon nucleophiles to N-protected a-amino aldehydes. This
chemistry has been used in numerous industrial processes,
including the production of marketed protease inhibitors
containing hydroxyethylene isosteres.[2] Owing to the inherent
incompatibility of the amine and aldehyde functionalities, the
protection of the nitrogen atom has been unavoidable in all of
these applications. Such protection disfavors undesired con-
densation reactions but increases the risk of racemization of
chiral amino aldehydes.[3] The recourse to protecting groups
has also produced complications in metal-mediated addition
reactions to chiral a-amino aldehydes, as delicate tuning of
the nitrogen substituent is required to selectively minimize
the competing governance of stereocontrol by either chela-
tion or nonchelation models.[4] Furthermore, the removal of
the protecting groups from the amino alcohol product is often
not trivial. This limitation is not restricted to the synthesis of
amino alcohols from amino aldehydes. Both olefin amino-
long-standing problems associated with the rapid formation
of complex nitrogen-containing molecules without protect-
ing-group manipulations. Undesired intermolecular iminium
ion formation from amphoteric aziridine aldehydes is disfa-
vored thermodynamically owing to the increase in ring strain
involved in such a process. Aziridine aldehydes can be
prepared from simple starting materials, such as a-amino
acids, and exist as stable dimers with the monomer/dimer
equilibrium lying towards the dimer in a variety of solvents
(Scheme 1). The addition of carbon nucleophiles to ampho-
Scheme 1. Monomer/dimer dynamics in amphoteric amino aldehydes.
hydroxylation[5] and more recent methods based on C H
teric aziridine aldehydes has been elusive until now. Herein
we describe how the curious structural preferences in the
course of aziridine aldehyde dimer dissociation enable the
protecting-group-free, stereoselective synthesis of complex
amino alcohols.
Initial investigations into the addition of carbon nucleo-
philes, such as Grignard and organolithium reagents, to
aziridine aldehyde dimers resulted in quantitative recovery of
the starting materials. We attribute this disappointing lack of
reactivity to unfavorable dimer dissociation under basic
conditions (Scheme 1). The deprotonated dimer is unreactive
towards nucleophiles. To access aldehyde reactivity, a means
for shifting the equilibrium to unveil the aldehyde function-
À
activation require subsequent deprotection steps.[6]
The rapid assembly of stereochemically complex b-amino
alcohol structures without recourse to the use of protecting
groups has been a long-standing challenge. We sought a
synthetic method that would not only deliver unprotected
amino alcohols but would also enable downstream diver-
gency. As direct progenitors of amino alcohols, unprotected
amino aldehydes can be viewed as key strategic building
blocks; however, there has been limited success with their
synthesis. One hundred years ago, Fischer attempted to
prepare glycinal, which was found to be unstable.[7] Many
years later, Myers et al. described autoprotection of the
amino functionality in a-amino aldehydes by treatment with
trifluoroacetic acid in methanol. The resulting hemiacetal
adducts are intriguing intermediates but are prone to self-
condensation above pH 5.[8]
ality in the presence of
a nucleophile was required
(Scheme 1). Our attention was directed to the use of protic
solvents and carbon–carbon bond-forming reactions medi-
ated by the water-tolerant indium reagents.[12] Gratifyingly,
the addition of allyl indium reagents to aziridine aldehyde
dimers was successful (Table 1). The scope of amino alcohol
formation was explored by using a variety of allyl bromides.
In all cases the chemical yields were high, with exclusive
production of the syn b-amino alcohols through g addition.
No undesired aziridine ring opening was observed in the
reaction, in contrast to the well-known scission of epoxides by
Our recent studies in the field of amphoteric mole-
cules[9–11] provided an opportunity to address some of the
[*] R. Hili, Prof. Dr. A. K. Yudin
Davenport Research Laboratories
Department of Chemistry, University of Toronto
80 St. George Street, Toronto, ON, M5S3H6 (Canada)
Fax: (+1)416-946-7676
allyl indium reagents under similar conditions.[13] A 1:1
E-mail: ayudin@chem.utoronto.ca
[12]
mixture of H2Oand THF
was optimal as the solvent in
[**] We thank the Natural Science and Engineering Research Council
(NSERC) and Canadian Institutes of Health Research (CIHR) for
financial support. R.H. is grateful to the NSERC for a postgraduate
fellowship.
terms of both the yield and the rate of the reaction, with the
syn diastereoisomers formed as the only detectable products.
No conversion was observed with other solvents, such as
trifluoroethanol, DMF, or anhydrous THF. The potential
utility of the 1,2-amino alcohol template is immediately
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4188
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4188 –4191

Angewandte
Chemie
Table 1: Amino alcohol synthesis with a variety of allyl bromides.[a]
Table 2: Stereoselective formation of sulfanyl amino alcohols.[a]
Entry
1
Allyl bromide
Product
Yield [%][b]
96
Entry
Allyl bromide
Product
Yield [%][b]
85[c]
1
2
95
2
3
4
5
6
92
88
91
79
92
3
96
4
5
96
81
6
7
90
87
[a] Upon completion of the allyl indium addition (as determined by TLC
analysis), the thiol (1 equiv) was added, and the reaction mixture was
heated at 608C for 1 h. [b] Yield of the isolated product. Bz=benzoyl.
8
86
[a] Reactions were performed on a 0.1 mmol scale (with respect to 1) at
0.1m in H2O/THF (1:1 v/v) with a dimer/In/allyl bromide ratio of
1:2.2:2.2. [b] Yield of the isolated product. [c] The reaction was
performed on a 1 mmol scale. TBDMS=tert-butyldimethylsilyl.
The complete diastereoselectivity observed during the
synthesis of unprotected amino alcohols called for a mech-
anistic explanation. At the outset, our efforts were aimed at
shifting the equilibrium towards the aldehyde, which was
projected to act as the electrophile (Scheme 1). However, is
the monomeric aldehyde the reactive species involved?
Evidence against the involvement of the monomeric aldehyde
is afforded by the reaction described in entry 8 of Table 1.
Only the N-allylated dimer was produced from the starting
allyl bromide; the N-allylated monomer was not observed.
That the monomeric species is not formed under these
reaction conditions is supported further by the lack of
crossover between two different aziridine aldehyde dimers.
A plausible mechanism that would operate in the absence of
complete dimer dissociation involves addition of the allyl
indium reagent to the latent aldehyde of type 2 (Scheme 2).
The latter is expected to chelate the allyl indium species to
facilitate the stereoselective delivery of the allyl group to the
aldehyde carbon atom. Quantum chemical calculations[18]
were performed to locate the transition-state structure for
the allylation reaction. It was found that the allyl indium(I)[19]
species chelates preferentially between the carbonyl oxygen
atom and the aziridine nitrogen atom to form the concave
transition state 3, which exhibits hydrogen bonding between
the incipient alkoxide and the hydroxy group of the carbinol-
amine. This interaction both directs the facial selectivity of the
allylation and stabilizes negative-charge buildup.[20] The
resulting product 4 is expected to undergo proton transfer
between the alkoxide and the hydroxy group of the carbinol-
amine to release the amino alcohol product 6 and the
evident. Protease inhibitors containing 1,2-amino alcohol
functionalities are not cleavable hydrolytically and are
recognized at the atomic level by their targets through
many different mechanisms.[14,15] Our versatile templates
provide an enabling technology for the synthesis of a wide
range of protease inhibitors.
To evaluate the possibility of one-pot operations that
result in useful stereodefined triads, we pursued the assembly
of sulfanyl amino alcohols. The sulfanyl amino alcohol motif
is found in protease inhibitors, such as nelfinavir.[16] A one-pot
allylation/nucleophilic ring-opening sequence was sought to
enable access to an unprotected S, N, Ostructural motif.
Upon completion of the allylation, benzenethiol was added to
the reaction mixture, which was then heated at 608C for 1 h.
The resulting products were obtained as single diastereoiso-
mers (Table 2).
We also developed a one-pot strategy that takes advant-
age of the unprotected aziridine and olefin functionalities
installed during the synthesis of the amino alcohol. An
important feature of unprotected aziridines is their resistance
to undesired oxidation to imines.[17] The addition of N-
bromosuccinimide (NBS) to the allylation reaction mixture at
08C led to the clean production of [3,5]bicycles as single
diastereoisomers (Table 3). This methodology enables the
facile construction of substituted pyrrolidinols equipped with
versatile aziridine substituents.
Angew. Chem. Int. Ed. 2008, 47, 4188 –4191
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4189

Communications
Table 3: Formation of pyrrolidinols from amphoteric amino aldehydes.[a]
ration of unprotected aziridines will facilitate the synthesis of
diverse syn amino alcohols, ubiquitous components of ther-
apeutically relevant protease inhibitors and other bioactive
compounds.
Experimental Section
Entry
1
Allyl bromide
Product
Yield [%][b]
79
Representative procedure (Table 1, entry 1): The dimer 1 (R1 = Ph;
294 mg, 1 mmol) was dissolved in a mixture of H2Oand THF (1:1 (v/
v); 10 mL) in a vial equipped with a magnetic stirring bar and a screw-
cap lid. Indium (255 mg, 2.2 mmol) was added, followed by allyl
bromide (186 ml, 2.2 mmol), and the reaction mixture was stirred at
room temperature for 1 h. Water and EtOAc were then added, and
the mixture was extracted three times with EtOAc. The combined
organic layers were dried over Na2SO4, filtered, and then concen-
trated under reduced pressure. The crude product (321.5 mg, 85%)
was isolated by flash column chromatography on silica gel (EtOAc;
Rf = 0.15). 1H NMR (400 MHz, CDCl3): d = 7.38–7.18 (m, 5H), 5.95–
5.81 (m, 1H), 5.20–5.10 (m, 2H), 3.61 (dd, J = 6.4, 12 Hz, 1H), 2.91 (d,
J = 3.2 Hz, 1H), 2.43–2.38 (m, 2H), 2.30 (br s, 1H), 2.20–1.40 ppm
(br s, 2H); 13C NMR (100 MHz, CDCl3): d = 139.4, 134.2, 128.8, 127.5,
125.9, 118.3, 71.1, 45.0, 40.7, 37.4 ppm; HRMS (ESI): m/z calcd for
C12H16NO: 190.1232 [M + H]+; found: 190.1234.
2
3
4
74
78
72
Received: December 17, 2007
Revised: February 18, 2008
Published online: April 25, 2008
[a] Upon completion of the allyl indium addition (as determined by TLC
analysis), NBS (1.1 equiv) was added to the reaction mixture at 08C, and
the mixture was stirred at 08C for 2 h. [b] Yield of the isolated product.
Keywords: amino alcohols · amino aldehydes ·
amphoteric molecules · protease inhibitors · synthetic methods
.
[1] For a review on the synthesis of amino alcohols, see: S. C.
Bergmeier, Tetrahedron 2000, 56, 2561 – 2576.
[2] K. Izawa, T. Onishi, Chem. Rev. 2006, 106, 2811 – 2827.
[3] a) R. W. Hoffmann, Synthesis 2006, 3531 – 3541; b) P. S. Baran,
T. J. Maimone, J. M. Richter, Nature 2007, 446, 404 – 408.
[4] M. Reetz, Angew. Chem. 1991, 103, 1559 – 1573; Angew. Chem.
Int. Ed. Engl. 1991, 30, 1531 – 1546.
[5] G. Li, H. T. Chang, K. B. Sharpless, Angew. Chem. 1996, 108,
449 – 452; Angew. Chem. Int. Ed. Engl. 1996, 35, 451 – 454.
[6] a) E. J. Alexanian, C. Lee, E. J. Sorensen, J. Am. Chem. Soc.
2005, 127, 7690 – 7691; b) K. J. Fraunhoffer, M. C. White, J. Am.
Chem. Soc. 2007, 129, 7274 – 7276.
[7] a) E. Fischer, Ber. Dtsch. Chem. Ges. 1908, 41, 956; b) E. Fischer,
Ber. Dtsch. Chem. Ges. 1908, 41, 1019.
[8] A. G. Myers, D. W. Kung, B. Zhong, J. Am. Chem. Soc. 2000, 122,
3236.
Scheme 2. Proposed mechanism of b-amino alcohol formation.
[9] A. K. Yudin, R. Hili, Chem. Eur. J. 2007, 13, 6538 – 6542.
[10] R. Hili, A. K. Yudin, J. Am. Chem. Soc. 2006, 128, 14772 – 14773.
[11] X. Li, A. K. Yudin, J. Am. Chem. Soc. 2007, 129, 14152 – 14153.
[12] a) C. J. Li, T.-H. Chan, Tetrahedron 1999, 55, 11149 – 11176; b) V.
Nair, S. Ros, C. N. Jayan, B. S. Pillai, Tetrahedron 2004, 60, 1959 –
1982.
[13] J. S. Yadav, S. Anjaneyulu, Md. M. Ahmed, B. V. S. Reddy,
Tetrahedron Lett. 2001, 42, 2557 – 2559.
[14] a) A. Krohn, S. Redshaw, J. C. Ritchie, B. J. Graves, M. H.
monomeric aziridine aldehyde 5. Upon release, 5 undergoes
rapid dimerization to reform the homochiral dimer adduct
with previously noted fidelity.[10]
In summary, a direct approach to unprotected syn amino
alcohols is possible through the indium-mediated addition of
carbon nucleophiles to readily available amphoteric amino
aldehydes. Efforts to shift the dimer/monomer equilibrium
revealed an important feature of the transformation: The
half-open species appears to be optimal with regard to both
reactivity and selectivity during amino alcohol formation. The
downstream utility of the resulting products has been
demonstrated in several one-pot operations, which led to
stereochemically complex scaffolds. This late-stage incorpo-
Hatada, J. Med. Chem. 1989, 32, 1532 – 1538; b) J. Dohnµlek, J.
ˇ
ˇ
ˇ
Hasek, J. Duskovµ, H. Petrokovµ, M. Hradilek, M. Soucek, J.
ˇ
Konvalinka, J. Brynda, J. Sedlµcek, M. Fµbry, Acta Crystallogr.
Sect. D 2001, 57, 472 – 476; c) E. Specker, J. Böttcher, H. Lilie, A.
Heine, A. Schoop, G. Müller, N. Gribenow, G. Kleber, Angew.
Chem. 2005, 117, 3200 – 3204; Angew. Chem. Int. Ed. 2005, 44,
3140 – 3144.
[15] C. E. Lee, E. K. Kick, J. A. Ellman, J. Am. Chem. Soc. 1998, 120,
9735 – 9747.
4190 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4188 –4191

Angewandte
Chemie
[16] K. Luzuriaga, M. McManus, L. Mofenson, P. Britto, B. Graham,
L. J. Sullivan, N. Engl. J. Med. 2004, 350, 2471 – 2480.
[17] a) M. Sasaki, A. K. Yudin, J. Am. Chem. Soc. 2003, 125, 14242 –
14243; b) G. Chen, M. Sasaki, X. Li, A. K. Yudin, J. Org. Chem.
2006, 71, 6067 – 6073.
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez,
J. A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[18] The optimized geometries of structures were calculated at the
B3LYP/STO-3G* level by using Gaussian03. The transition
state was located by employing the QST2 routine with optimi-
zation and frequency analysis (NoRaman) performed at the
B3LYP/STO-3G* level. See the Supporting Information for
output files. Gaussian03, RevisionD.01, M. J. Frisch, G. W.
Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheese-
man, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
[19] T. H. Chan, Y. Yang, J. Am. Chem. Soc. 1999, 121, 3228 – 3229.
[20] Hydrogen bonding between a developing alkoxide and an
appropriate donor group is known to act as a control element
that leads to lower activation energies; see: C. Allemann, R.
Gordillo, F. R. Clemente, P. H. Y. Cheong, K. N. Houk, Acc.
Chem. Res. 2004, 37, 558 – 569.
Angew. Chem. Int. Ed. 2008, 47, 4188 –4191
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4191